Upload 15 files

.gitattributes

@@ -41,3 +41,4 @@ gpt4_alpaca/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 oasst1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 open_orca/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 science_literature/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+sharegpt/tokenizer.json filter=lfs diff=lfs merge=lfs -text

sharegpt/README.md

@@ -0,0 +1,69 @@
+---
+base_model: /cpfs01/user/zhangyiqun/models/gemma-2-2b-it
+library_name: peft
+license: other
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: sharegpt
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# sharegpt
+
+This model is a fine-tuned version of [/cpfs01/user/zhangyiqun/models/gemma-2-2b-it](https://huggingface.co//cpfs01/user/zhangyiqun/models/gemma-2-2b-it) on the sharegpt dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7986
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 2
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 32
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1.0
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.7886        | 0.3010 | 1000 | 0.8210          |
+| 0.7502        | 0.6020 | 2000 | 0.8055          |
+| 0.7516        | 0.9029 | 3000 | 0.7988          |
+
+
+### Framework versions
+
+- PEFT 0.12.0
+- Transformers 4.44.2
+- Pytorch 2.3.0+cu121
+- Datasets 3.0.1
+- Tokenizers 0.19.1

sharegpt/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/cpfs01/user/zhangyiqun/models/gemma-2-2b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "down_proj",
+    "k_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

sharegpt/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00ff1d647ea7ecd6f0fc32382ae325696751460696bd2ef4ab763abb28fdcbef
+size 41581360

sharegpt/all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 0.999849510910459,
+    "eval_loss": 0.7985649704933167,
+    "eval_runtime": 142.8485,
+    "eval_samples_per_second": 39.174,
+    "eval_steps_per_second": 4.9,
+    "total_flos": 3.156428776263385e+18,
+    "train_loss": 0.7673604141007986,
+    "train_runtime": 8289.1597,
+    "train_samples_per_second": 12.826,
+    "train_steps_per_second": 0.401
+}

sharegpt/eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 0.999849510910459,
+    "eval_loss": 0.7985649704933167,
+    "eval_runtime": 142.8485,
+    "eval_samples_per_second": 39.174,
+    "eval_steps_per_second": 4.9
+}

sharegpt/special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

sharegpt/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357

sharegpt/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003

sharegpt/tokenizer_config.json

@@ -0,0 +1,2015 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

sharegpt/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.999849510910459,
+    "total_flos": 3.156428776263385e+18,
+    "train_loss": 0.7673604141007986,
+    "train_runtime": 8289.1597,
+    "train_samples_per_second": 12.826,
+    "train_steps_per_second": 0.401
+}

sharegpt/trainer_log.jsonl

@@ -0,0 +1,336 @@
+{"current_steps": 10, "total_steps": 3322, "loss": 1.0909, "learning_rate": 6.006006006006006e-06, "epoch": 0.0030097817908201654, "percentage": 0.3, "elapsed_time": "0:00:24", "remaining_time": "2:17:50"}
+{"current_steps": 20, "total_steps": 3322, "loss": 1.0813, "learning_rate": 1.2012012012012012e-05, "epoch": 0.006019563581640331, "percentage": 0.6, "elapsed_time": "0:00:50", "remaining_time": "2:18:46"}
+{"current_steps": 30, "total_steps": 3322, "loss": 1.051, "learning_rate": 1.801801801801802e-05, "epoch": 0.009029345372460496, "percentage": 0.9, "elapsed_time": "0:01:14", "remaining_time": "2:15:45"}
+{"current_steps": 40, "total_steps": 3322, "loss": 0.9406, "learning_rate": 2.4024024024024024e-05, "epoch": 0.012039127163280662, "percentage": 1.2, "elapsed_time": "0:01:39", "remaining_time": "2:16:12"}
+{"current_steps": 50, "total_steps": 3322, "loss": 0.9232, "learning_rate": 3.0030030030030033e-05, "epoch": 0.015048908954100828, "percentage": 1.51, "elapsed_time": "0:02:05", "remaining_time": "2:16:33"}
+{"current_steps": 60, "total_steps": 3322, "loss": 0.8762, "learning_rate": 3.603603603603604e-05, "epoch": 0.01805869074492099, "percentage": 1.81, "elapsed_time": "0:02:28", "remaining_time": "2:14:40"}
+{"current_steps": 70, "total_steps": 3322, "loss": 0.8496, "learning_rate": 4.204204204204204e-05, "epoch": 0.021068472535741158, "percentage": 2.11, "elapsed_time": "0:02:52", "remaining_time": "2:13:15"}
+{"current_steps": 80, "total_steps": 3322, "loss": 0.8424, "learning_rate": 4.804804804804805e-05, "epoch": 0.024078254326561323, "percentage": 2.41, "elapsed_time": "0:03:15", "remaining_time": "2:12:08"}
+{"current_steps": 90, "total_steps": 3322, "loss": 0.8601, "learning_rate": 5.405405405405406e-05, "epoch": 0.02708803611738149, "percentage": 2.71, "elapsed_time": "0:03:38", "remaining_time": "2:10:57"}
+{"current_steps": 100, "total_steps": 3322, "loss": 0.8179, "learning_rate": 6.0060060060060066e-05, "epoch": 0.030097817908201655, "percentage": 3.01, "elapsed_time": "0:04:01", "remaining_time": "2:09:57"}
+{"current_steps": 110, "total_steps": 3322, "loss": 0.8094, "learning_rate": 6.606606606606607e-05, "epoch": 0.03310759969902182, "percentage": 3.31, "elapsed_time": "0:04:27", "remaining_time": "2:10:02"}
+{"current_steps": 120, "total_steps": 3322, "loss": 0.8096, "learning_rate": 7.207207207207208e-05, "epoch": 0.03611738148984198, "percentage": 3.61, "elapsed_time": "0:04:52", "remaining_time": "2:10:04"}
+{"current_steps": 130, "total_steps": 3322, "loss": 0.8195, "learning_rate": 7.807807807807808e-05, "epoch": 0.03912716328066215, "percentage": 3.91, "elapsed_time": "0:05:15", "remaining_time": "2:09:11"}
+{"current_steps": 140, "total_steps": 3322, "loss": 0.8184, "learning_rate": 8.408408408408409e-05, "epoch": 0.042136945071482315, "percentage": 4.21, "elapsed_time": "0:05:39", "remaining_time": "2:08:39"}
+{"current_steps": 150, "total_steps": 3322, "loss": 0.8, "learning_rate": 9.009009009009009e-05, "epoch": 0.045146726862302484, "percentage": 4.52, "elapsed_time": "0:06:04", "remaining_time": "2:08:23"}
+{"current_steps": 160, "total_steps": 3322, "loss": 0.7739, "learning_rate": 9.60960960960961e-05, "epoch": 0.04815650865312265, "percentage": 4.82, "elapsed_time": "0:06:28", "remaining_time": "2:07:51"}
+{"current_steps": 170, "total_steps": 3322, "loss": 0.8172, "learning_rate": 0.00010210210210210212, "epoch": 0.051166290443942816, "percentage": 5.12, "elapsed_time": "0:06:51", "remaining_time": "2:07:14"}
+{"current_steps": 180, "total_steps": 3322, "loss": 0.8181, "learning_rate": 0.00010810810810810812, "epoch": 0.05417607223476298, "percentage": 5.42, "elapsed_time": "0:07:16", "remaining_time": "2:06:58"}
+{"current_steps": 190, "total_steps": 3322, "loss": 0.8032, "learning_rate": 0.00011411411411411413, "epoch": 0.05718585402558315, "percentage": 5.72, "elapsed_time": "0:07:40", "remaining_time": "2:06:30"}
+{"current_steps": 200, "total_steps": 3322, "loss": 0.7775, "learning_rate": 0.00012012012012012013, "epoch": 0.06019563581640331, "percentage": 6.02, "elapsed_time": "0:08:05", "remaining_time": "2:06:25"}
+{"current_steps": 210, "total_steps": 3322, "loss": 0.821, "learning_rate": 0.00012612612612612612, "epoch": 0.06320541760722348, "percentage": 6.32, "elapsed_time": "0:08:29", "remaining_time": "2:05:50"}
+{"current_steps": 220, "total_steps": 3322, "loss": 0.7892, "learning_rate": 0.00013213213213213214, "epoch": 0.06621519939804364, "percentage": 6.62, "elapsed_time": "0:08:52", "remaining_time": "2:05:05"}
+{"current_steps": 230, "total_steps": 3322, "loss": 0.7932, "learning_rate": 0.00013813813813813813, "epoch": 0.0692249811888638, "percentage": 6.92, "elapsed_time": "0:09:15", "remaining_time": "2:04:26"}
+{"current_steps": 240, "total_steps": 3322, "loss": 0.7993, "learning_rate": 0.00014414414414414415, "epoch": 0.07223476297968397, "percentage": 7.22, "elapsed_time": "0:09:38", "remaining_time": "2:03:55"}
+{"current_steps": 250, "total_steps": 3322, "loss": 0.8281, "learning_rate": 0.00015015015015015014, "epoch": 0.07524454477050414, "percentage": 7.53, "elapsed_time": "0:10:02", "remaining_time": "2:03:22"}
+{"current_steps": 260, "total_steps": 3322, "loss": 0.7695, "learning_rate": 0.00015615615615615616, "epoch": 0.0782543265613243, "percentage": 7.83, "elapsed_time": "0:10:25", "remaining_time": "2:02:49"}
+{"current_steps": 270, "total_steps": 3322, "loss": 0.7863, "learning_rate": 0.00016216216216216218, "epoch": 0.08126410835214447, "percentage": 8.13, "elapsed_time": "0:10:49", "remaining_time": "2:02:23"}
+{"current_steps": 280, "total_steps": 3322, "loss": 0.7616, "learning_rate": 0.00016816816816816817, "epoch": 0.08427389014296463, "percentage": 8.43, "elapsed_time": "0:11:13", "remaining_time": "2:02:00"}
+{"current_steps": 290, "total_steps": 3322, "loss": 0.7707, "learning_rate": 0.0001741741741741742, "epoch": 0.0872836719337848, "percentage": 8.73, "elapsed_time": "0:11:38", "remaining_time": "2:01:45"}
+{"current_steps": 300, "total_steps": 3322, "loss": 0.7744, "learning_rate": 0.00018018018018018018, "epoch": 0.09029345372460497, "percentage": 9.03, "elapsed_time": "0:12:01", "remaining_time": "2:01:12"}
+{"current_steps": 310, "total_steps": 3322, "loss": 0.743, "learning_rate": 0.0001861861861861862, "epoch": 0.09330323551542513, "percentage": 9.33, "elapsed_time": "0:12:25", "remaining_time": "2:00:40"}
+{"current_steps": 320, "total_steps": 3322, "loss": 0.8089, "learning_rate": 0.0001921921921921922, "epoch": 0.0963130173062453, "percentage": 9.63, "elapsed_time": "0:12:49", "remaining_time": "2:00:16"}
+{"current_steps": 330, "total_steps": 3322, "loss": 0.8012, "learning_rate": 0.0001981981981981982, "epoch": 0.09932279909706546, "percentage": 9.93, "elapsed_time": "0:13:12", "remaining_time": "1:59:49"}
+{"current_steps": 340, "total_steps": 3322, "loss": 0.7843, "learning_rate": 0.00019999729347501484, "epoch": 0.10233258088788563, "percentage": 10.23, "elapsed_time": "0:13:36", "remaining_time": "1:59:25"}
+{"current_steps": 350, "total_steps": 3322, "loss": 0.7992, "learning_rate": 0.0001999840373787939, "epoch": 0.1053423626787058, "percentage": 10.54, "elapsed_time": "0:14:00", "remaining_time": "1:58:58"}
+{"current_steps": 360, "total_steps": 3322, "loss": 0.772, "learning_rate": 0.0001999597360570722, "epoch": 0.10835214446952596, "percentage": 10.84, "elapsed_time": "0:14:24", "remaining_time": "1:58:34"}
+{"current_steps": 370, "total_steps": 3322, "loss": 0.7511, "learning_rate": 0.0001999243921944139, "epoch": 0.11136192626034612, "percentage": 11.14, "elapsed_time": "0:14:48", "remaining_time": "1:58:12"}
+{"current_steps": 380, "total_steps": 3322, "loss": 0.7735, "learning_rate": 0.00019987800969525164, "epoch": 0.1143717080511663, "percentage": 11.44, "elapsed_time": "0:15:12", "remaining_time": "1:57:41"}
+{"current_steps": 390, "total_steps": 3322, "loss": 0.7729, "learning_rate": 0.00019982059368345496, "epoch": 0.11738148984198646, "percentage": 11.74, "elapsed_time": "0:15:35", "remaining_time": "1:57:10"}
+{"current_steps": 400, "total_steps": 3322, "loss": 0.795, "learning_rate": 0.00019975215050176433, "epoch": 0.12039127163280662, "percentage": 12.04, "elapsed_time": "0:15:58", "remaining_time": "1:56:45"}
+{"current_steps": 410, "total_steps": 3322, "loss": 0.7671, "learning_rate": 0.00019967268771109035, "epoch": 0.12340105342362678, "percentage": 12.34, "elapsed_time": "0:16:22", "remaining_time": "1:56:18"}
+{"current_steps": 420, "total_steps": 3322, "loss": 0.7491, "learning_rate": 0.00019958221408967875, "epoch": 0.12641083521444696, "percentage": 12.64, "elapsed_time": "0:16:46", "remaining_time": "1:55:54"}
+{"current_steps": 430, "total_steps": 3322, "loss": 0.7766, "learning_rate": 0.00019948073963214043, "epoch": 0.1294206170052671, "percentage": 12.94, "elapsed_time": "0:17:11", "remaining_time": "1:55:34"}
+{"current_steps": 440, "total_steps": 3322, "loss": 0.7904, "learning_rate": 0.00019936827554834738, "epoch": 0.13243039879608728, "percentage": 13.25, "elapsed_time": "0:17:35", "remaining_time": "1:55:13"}
+{"current_steps": 450, "total_steps": 3322, "loss": 0.7816, "learning_rate": 0.00019924483426219452, "epoch": 0.13544018058690746, "percentage": 13.55, "elapsed_time": "0:17:58", "remaining_time": "1:54:41"}
+{"current_steps": 460, "total_steps": 3322, "loss": 0.7982, "learning_rate": 0.00019911042941022695, "epoch": 0.1384499623777276, "percentage": 13.85, "elapsed_time": "0:18:22", "remaining_time": "1:54:16"}
+{"current_steps": 470, "total_steps": 3322, "loss": 0.793, "learning_rate": 0.00019896507584013376, "epoch": 0.14145974416854779, "percentage": 14.15, "elapsed_time": "0:18:45", "remaining_time": "1:53:50"}
+{"current_steps": 480, "total_steps": 3322, "loss": 0.745, "learning_rate": 0.00019880878960910772, "epoch": 0.14446952595936793, "percentage": 14.45, "elapsed_time": "0:19:10", "remaining_time": "1:53:30"}
+{"current_steps": 490, "total_steps": 3322, "loss": 0.765, "learning_rate": 0.00019864158798207137, "epoch": 0.1474793077501881, "percentage": 14.75, "elapsed_time": "0:19:35", "remaining_time": "1:53:13"}
+{"current_steps": 500, "total_steps": 3322, "loss": 0.7494, "learning_rate": 0.0001984634894297699, "epoch": 0.1504890895410083, "percentage": 15.05, "elapsed_time": "0:19:59", "remaining_time": "1:52:47"}
+{"current_steps": 510, "total_steps": 3322, "loss": 0.7906, "learning_rate": 0.00019827451362673052, "epoch": 0.15349887133182843, "percentage": 15.35, "elapsed_time": "0:20:23", "remaining_time": "1:52:24"}
+{"current_steps": 520, "total_steps": 3322, "loss": 0.7991, "learning_rate": 0.00019807468144908928, "epoch": 0.1565086531226486, "percentage": 15.65, "elapsed_time": "0:20:47", "remaining_time": "1:52:00"}
+{"current_steps": 530, "total_steps": 3322, "loss": 0.7593, "learning_rate": 0.00019786401497228466, "epoch": 0.1595184349134688, "percentage": 15.95, "elapsed_time": "0:21:10", "remaining_time": "1:51:33"}
+{"current_steps": 540, "total_steps": 3322, "loss": 0.7966, "learning_rate": 0.00019764253746861886, "epoch": 0.16252821670428894, "percentage": 16.26, "elapsed_time": "0:21:34", "remaining_time": "1:51:06"}
+{"current_steps": 550, "total_steps": 3322, "loss": 0.7525, "learning_rate": 0.00019741027340468715, "epoch": 0.1655379984951091, "percentage": 16.56, "elapsed_time": "0:21:57", "remaining_time": "1:50:41"}
+{"current_steps": 560, "total_steps": 3322, "loss": 0.7706, "learning_rate": 0.00019716724843867487, "epoch": 0.16854778028592926, "percentage": 16.86, "elapsed_time": "0:22:21", "remaining_time": "1:50:17"}
+{"current_steps": 570, "total_steps": 3322, "loss": 0.7683, "learning_rate": 0.000196913489417523, "epoch": 0.17155756207674944, "percentage": 17.16, "elapsed_time": "0:22:44", "remaining_time": "1:49:49"}
+{"current_steps": 580, "total_steps": 3322, "loss": 0.821, "learning_rate": 0.00019664902437396245, "epoch": 0.1745673438675696, "percentage": 17.46, "elapsed_time": "0:23:07", "remaining_time": "1:49:18"}
+{"current_steps": 590, "total_steps": 3322, "loss": 0.7744, "learning_rate": 0.00019637388252341715, "epoch": 0.17757712565838976, "percentage": 17.76, "elapsed_time": "0:23:29", "remaining_time": "1:48:47"}
+{"current_steps": 600, "total_steps": 3322, "loss": 0.7607, "learning_rate": 0.00019608809426077678, "epoch": 0.18058690744920994, "percentage": 18.06, "elapsed_time": "0:23:54", "remaining_time": "1:48:26"}
+{"current_steps": 610, "total_steps": 3322, "loss": 0.7707, "learning_rate": 0.000195791691157039, "epoch": 0.1835966892400301, "percentage": 18.36, "elapsed_time": "0:24:18", "remaining_time": "1:48:04"}
+{"current_steps": 620, "total_steps": 3322, "loss": 0.7487, "learning_rate": 0.00019548470595582166, "epoch": 0.18660647103085026, "percentage": 18.66, "elapsed_time": "0:24:41", "remaining_time": "1:47:37"}
+{"current_steps": 630, "total_steps": 3322, "loss": 0.7653, "learning_rate": 0.00019516717256974592, "epoch": 0.18961625282167044, "percentage": 18.96, "elapsed_time": "0:25:05", "remaining_time": "1:47:13"}
+{"current_steps": 640, "total_steps": 3322, "loss": 0.7918, "learning_rate": 0.00019483912607668965, "epoch": 0.1926260346124906, "percentage": 19.27, "elapsed_time": "0:25:29", "remaining_time": "1:46:48"}
+{"current_steps": 650, "total_steps": 3322, "loss": 0.8022, "learning_rate": 0.00019450060271591243, "epoch": 0.19563581640331076, "percentage": 19.57, "elapsed_time": "0:25:51", "remaining_time": "1:46:19"}
+{"current_steps": 660, "total_steps": 3322, "loss": 0.7548, "learning_rate": 0.0001941516398840524, "epoch": 0.1986455981941309, "percentage": 19.87, "elapsed_time": "0:26:16", "remaining_time": "1:45:56"}
+{"current_steps": 670, "total_steps": 3322, "loss": 0.7903, "learning_rate": 0.00019379227613099473, "epoch": 0.2016553799849511, "percentage": 20.17, "elapsed_time": "0:26:39", "remaining_time": "1:45:30"}
+{"current_steps": 680, "total_steps": 3322, "loss": 0.7917, "learning_rate": 0.00019342255115561337, "epoch": 0.20466516177577126, "percentage": 20.47, "elapsed_time": "0:27:02", "remaining_time": "1:45:03"}
+{"current_steps": 690, "total_steps": 3322, "loss": 0.7879, "learning_rate": 0.00019304250580138524, "epoch": 0.2076749435665914, "percentage": 20.77, "elapsed_time": "0:27:24", "remaining_time": "1:44:33"}
+{"current_steps": 700, "total_steps": 3322, "loss": 0.8014, "learning_rate": 0.0001926521820518784, "epoch": 0.2106847253574116, "percentage": 21.07, "elapsed_time": "0:27:48", "remaining_time": "1:44:09"}
+{"current_steps": 710, "total_steps": 3322, "loss": 0.7403, "learning_rate": 0.00019225162302611412, "epoch": 0.21369450714823177, "percentage": 21.37, "elapsed_time": "0:28:12", "remaining_time": "1:43:46"}
+{"current_steps": 720, "total_steps": 3322, "loss": 0.7795, "learning_rate": 0.00019184087297380344, "epoch": 0.21670428893905191, "percentage": 21.67, "elapsed_time": "0:28:36", "remaining_time": "1:43:23"}
+{"current_steps": 730, "total_steps": 3322, "loss": 0.8047, "learning_rate": 0.000191419977270459, "epoch": 0.2197140707298721, "percentage": 21.97, "elapsed_time": "0:29:00", "remaining_time": "1:42:58"}
+{"current_steps": 740, "total_steps": 3322, "loss": 0.7786, "learning_rate": 0.0001909889824123824, "epoch": 0.22272385252069224, "percentage": 22.28, "elapsed_time": "0:29:23", "remaining_time": "1:42:33"}
+{"current_steps": 750, "total_steps": 3322, "loss": 0.7352, "learning_rate": 0.00019054793601152773, "epoch": 0.22573363431151242, "percentage": 22.58, "elapsed_time": "0:29:47", "remaining_time": "1:42:09"}
+{"current_steps": 760, "total_steps": 3322, "loss": 0.778, "learning_rate": 0.0001900968867902419, "epoch": 0.2287434161023326, "percentage": 22.88, "elapsed_time": "0:30:11", "remaining_time": "1:41:47"}
+{"current_steps": 770, "total_steps": 3322, "loss": 0.7776, "learning_rate": 0.00018963588457588228, "epoch": 0.23175319789315274, "percentage": 23.18, "elapsed_time": "0:30:35", "remaining_time": "1:41:24"}
+{"current_steps": 780, "total_steps": 3322, "loss": 0.7841, "learning_rate": 0.00018916498029531223, "epoch": 0.23476297968397292, "percentage": 23.48, "elapsed_time": "0:31:00", "remaining_time": "1:41:02"}
+{"current_steps": 790, "total_steps": 3322, "loss": 0.7628, "learning_rate": 0.00018868422596927535, "epoch": 0.23777276147479307, "percentage": 23.78, "elapsed_time": "0:31:23", "remaining_time": "1:40:37"}
+{"current_steps": 800, "total_steps": 3322, "loss": 0.7403, "learning_rate": 0.00018819367470664862, "epoch": 0.24078254326561324, "percentage": 24.08, "elapsed_time": "0:31:48", "remaining_time": "1:40:15"}
+{"current_steps": 810, "total_steps": 3322, "loss": 0.7581, "learning_rate": 0.00018769338069857548, "epoch": 0.24379232505643342, "percentage": 24.38, "elapsed_time": "0:32:11", "remaining_time": "1:39:49"}
+{"current_steps": 820, "total_steps": 3322, "loss": 0.7914, "learning_rate": 0.00018718339921247945, "epoch": 0.24680210684725357, "percentage": 24.68, "elapsed_time": "0:32:35", "remaining_time": "1:39:26"}
+{"current_steps": 830, "total_steps": 3322, "loss": 0.7953, "learning_rate": 0.0001866637865859586, "epoch": 0.24981188863807374, "percentage": 24.98, "elapsed_time": "0:32:58", "remaining_time": "1:38:59"}
+{"current_steps": 840, "total_steps": 3322, "loss": 0.7576, "learning_rate": 0.00018613460022056215, "epoch": 0.2528216704288939, "percentage": 25.29, "elapsed_time": "0:33:21", "remaining_time": "1:38:34"}
+{"current_steps": 850, "total_steps": 3322, "loss": 0.7508, "learning_rate": 0.000185595898575449, "epoch": 0.2558314522197141, "percentage": 25.59, "elapsed_time": "0:33:45", "remaining_time": "1:38:11"}
+{"current_steps": 860, "total_steps": 3322, "loss": 0.7332, "learning_rate": 0.00018504774116093008, "epoch": 0.2588412340105342, "percentage": 25.89, "elapsed_time": "0:34:08", "remaining_time": "1:37:44"}
+{"current_steps": 870, "total_steps": 3322, "loss": 0.756, "learning_rate": 0.00018449018853189403, "epoch": 0.2618510158013544, "percentage": 26.19, "elapsed_time": "0:34:31", "remaining_time": "1:37:18"}
+{"current_steps": 880, "total_steps": 3322, "loss": 0.7776, "learning_rate": 0.0001839233022811179, "epoch": 0.26486079759217457, "percentage": 26.49, "elapsed_time": "0:34:55", "remaining_time": "1:36:55"}
+{"current_steps": 890, "total_steps": 3322, "loss": 0.7771, "learning_rate": 0.00018334714503246273, "epoch": 0.26787057938299474, "percentage": 26.79, "elapsed_time": "0:35:19", "remaining_time": "1:36:31"}
+{"current_steps": 900, "total_steps": 3322, "loss": 0.7544, "learning_rate": 0.00018276178043395586, "epoch": 0.2708803611738149, "percentage": 27.09, "elapsed_time": "0:35:43", "remaining_time": "1:36:07"}
+{"current_steps": 910, "total_steps": 3322, "loss": 0.7946, "learning_rate": 0.00018216727315075945, "epoch": 0.27389014296463504, "percentage": 27.39, "elapsed_time": "0:36:06", "remaining_time": "1:35:42"}
+{"current_steps": 920, "total_steps": 3322, "loss": 0.8202, "learning_rate": 0.00018156368885802695, "epoch": 0.2768999247554552, "percentage": 27.69, "elapsed_time": "0:36:28", "remaining_time": "1:35:14"}
+{"current_steps": 930, "total_steps": 3322, "loss": 0.7823, "learning_rate": 0.00018095109423364817, "epoch": 0.2799097065462754, "percentage": 28.0, "elapsed_time": "0:36:52", "remaining_time": "1:34:50"}
+{"current_steps": 940, "total_steps": 3322, "loss": 0.7637, "learning_rate": 0.0001803295569508832, "epoch": 0.28291948833709557, "percentage": 28.3, "elapsed_time": "0:37:16", "remaining_time": "1:34:27"}
+{"current_steps": 950, "total_steps": 3322, "loss": 0.768, "learning_rate": 0.0001796991456708866, "epoch": 0.28592927012791575, "percentage": 28.6, "elapsed_time": "0:37:39", "remaining_time": "1:34:02"}
+{"current_steps": 960, "total_steps": 3322, "loss": 0.7492, "learning_rate": 0.0001790599300351225, "epoch": 0.28893905191873587, "percentage": 28.9, "elapsed_time": "0:38:04", "remaining_time": "1:33:39"}
+{"current_steps": 970, "total_steps": 3322, "loss": 0.758, "learning_rate": 0.00017841198065767107, "epoch": 0.29194883370955604, "percentage": 29.2, "elapsed_time": "0:38:28", "remaining_time": "1:33:16"}
+{"current_steps": 980, "total_steps": 3322, "loss": 0.7739, "learning_rate": 0.00017775536911742806, "epoch": 0.2949586155003762, "percentage": 29.5, "elapsed_time": "0:38:51", "remaining_time": "1:32:52"}
+{"current_steps": 990, "total_steps": 3322, "loss": 0.7692, "learning_rate": 0.00017709016795019742, "epoch": 0.2979683972911964, "percentage": 29.8, "elapsed_time": "0:39:15", "remaining_time": "1:32:27"}
+{"current_steps": 1000, "total_steps": 3322, "loss": 0.7886, "learning_rate": 0.00017641645064067816, "epoch": 0.3009781790820166, "percentage": 30.1, "elapsed_time": "0:39:38", "remaining_time": "1:32:02"}
+{"current_steps": 1000, "total_steps": 3322, "eval_loss": 0.8210044503211975, "epoch": 0.3009781790820166, "percentage": 30.1, "elapsed_time": "0:42:01", "remaining_time": "1:37:34"}
+{"current_steps": 1010, "total_steps": 3322, "loss": 0.7595, "learning_rate": 0.0001757342916143466, "epoch": 0.3039879608728367, "percentage": 30.4, "elapsed_time": "0:42:25", "remaining_time": "1:37:07"}
+{"current_steps": 1020, "total_steps": 3322, "loss": 0.7821, "learning_rate": 0.00017504376622923465, "epoch": 0.30699774266365687, "percentage": 30.7, "elapsed_time": "0:42:49", "remaining_time": "1:36:39"}
+{"current_steps": 1030, "total_steps": 3322, "loss": 0.7982, "learning_rate": 0.00017434495076760483, "epoch": 0.31000752445447705, "percentage": 31.01, "elapsed_time": "0:43:12", "remaining_time": "1:36:09"}
+{"current_steps": 1040, "total_steps": 3322, "loss": 0.7422, "learning_rate": 0.00017363792242752353, "epoch": 0.3130173062452972, "percentage": 31.31, "elapsed_time": "0:43:36", "remaining_time": "1:35:41"}
+{"current_steps": 1050, "total_steps": 3322, "loss": 0.7425, "learning_rate": 0.000172922759314333, "epoch": 0.3160270880361174, "percentage": 31.61, "elapsed_time": "0:44:00", "remaining_time": "1:35:13"}
+{"current_steps": 1060, "total_steps": 3322, "loss": 0.7392, "learning_rate": 0.0001721995404320228, "epoch": 0.3190368698269376, "percentage": 31.91, "elapsed_time": "0:44:24", "remaining_time": "1:34:45"}
+{"current_steps": 1070, "total_steps": 3322, "loss": 0.7913, "learning_rate": 0.0001714683456745026, "epoch": 0.3220466516177577, "percentage": 32.21, "elapsed_time": "0:44:48", "remaining_time": "1:34:17"}
+{"current_steps": 1080, "total_steps": 3322, "loss": 0.7368, "learning_rate": 0.00017072925581677594, "epoch": 0.32505643340857787, "percentage": 32.51, "elapsed_time": "0:45:11", "remaining_time": "1:33:49"}
+{"current_steps": 1090, "total_steps": 3322, "loss": 0.7938, "learning_rate": 0.0001699823525060174, "epoch": 0.32806621519939805, "percentage": 32.81, "elapsed_time": "0:45:37", "remaining_time": "1:33:24"}
+{"current_steps": 1100, "total_steps": 3322, "loss": 0.7672, "learning_rate": 0.00016922771825255263, "epoch": 0.3310759969902182, "percentage": 33.11, "elapsed_time": "0:46:01", "remaining_time": "1:32:58"}
+{"current_steps": 1110, "total_steps": 3322, "loss": 0.7971, "learning_rate": 0.0001684654364207438, "epoch": 0.3340857787810384, "percentage": 33.41, "elapsed_time": "0:46:24", "remaining_time": "1:32:29"}
+{"current_steps": 1120, "total_steps": 3322, "loss": 0.7283, "learning_rate": 0.00016769559121978026, "epoch": 0.3370955605718585, "percentage": 33.71, "elapsed_time": "0:46:48", "remaining_time": "1:32:02"}
+{"current_steps": 1130, "total_steps": 3322, "loss": 0.7405, "learning_rate": 0.0001669182676943757, "epoch": 0.3401053423626787, "percentage": 34.02, "elapsed_time": "0:47:12", "remaining_time": "1:31:34"}
+{"current_steps": 1140, "total_steps": 3322, "loss": 0.7821, "learning_rate": 0.0001661335517153737, "epoch": 0.3431151241534989, "percentage": 34.32, "elapsed_time": "0:47:36", "remaining_time": "1:31:07"}
+{"current_steps": 1150, "total_steps": 3322, "loss": 0.7392, "learning_rate": 0.00016534152997026125, "epoch": 0.34612490594431905, "percentage": 34.62, "elapsed_time": "0:48:00", "remaining_time": "1:30:40"}
+{"current_steps": 1160, "total_steps": 3322, "loss": 0.7928, "learning_rate": 0.00016454228995359252, "epoch": 0.3491346877351392, "percentage": 34.92, "elapsed_time": "0:48:23", "remaining_time": "1:30:11"}
+{"current_steps": 1170, "total_steps": 3322, "loss": 0.7542, "learning_rate": 0.00016373591995732338, "epoch": 0.35214446952595935, "percentage": 35.22, "elapsed_time": "0:48:47", "remaining_time": "1:29:45"}
+{"current_steps": 1180, "total_steps": 3322, "loss": 0.7766, "learning_rate": 0.0001629225090610577, "epoch": 0.3551542513167795, "percentage": 35.52, "elapsed_time": "0:49:11", "remaining_time": "1:29:18"}
+{"current_steps": 1190, "total_steps": 3322, "loss": 0.7528, "learning_rate": 0.00016210214712220687, "epoch": 0.3581640331075997, "percentage": 35.82, "elapsed_time": "0:49:35", "remaining_time": "1:28:51"}
+{"current_steps": 1200, "total_steps": 3322, "loss": 0.7874, "learning_rate": 0.00016127492476606308, "epoch": 0.3611738148984199, "percentage": 36.12, "elapsed_time": "0:49:59", "remaining_time": "1:28:24"}
+{"current_steps": 1210, "total_steps": 3322, "loss": 0.7599, "learning_rate": 0.00016044093337578815, "epoch": 0.36418359668924005, "percentage": 36.42, "elapsed_time": "0:50:23", "remaining_time": "1:27:57"}
+{"current_steps": 1220, "total_steps": 3322, "loss": 0.7707, "learning_rate": 0.00015960026508231824, "epoch": 0.3671933784800602, "percentage": 36.72, "elapsed_time": "0:50:47", "remaining_time": "1:27:31"}
+{"current_steps": 1230, "total_steps": 3322, "loss": 0.8102, "learning_rate": 0.00015875301275418638, "epoch": 0.37020316027088035, "percentage": 37.03, "elapsed_time": "0:51:10", "remaining_time": "1:27:02"}
+{"current_steps": 1240, "total_steps": 3322, "loss": 0.7388, "learning_rate": 0.00015789926998726315, "epoch": 0.3732129420617005, "percentage": 37.33, "elapsed_time": "0:51:34", "remaining_time": "1:26:36"}
+{"current_steps": 1250, "total_steps": 3322, "loss": 0.7707, "learning_rate": 0.00015703913109441713, "epoch": 0.3762227238525207, "percentage": 37.63, "elapsed_time": "0:51:57", "remaining_time": "1:26:07"}
+{"current_steps": 1260, "total_steps": 3322, "loss": 0.7444, "learning_rate": 0.0001561726910950962, "epoch": 0.3792325056433409, "percentage": 37.93, "elapsed_time": "0:52:21", "remaining_time": "1:25:40"}
+{"current_steps": 1270, "total_steps": 3322, "loss": 0.7838, "learning_rate": 0.00015530004570483093, "epoch": 0.382242287434161, "percentage": 38.23, "elapsed_time": "0:52:44", "remaining_time": "1:25:13"}
+{"current_steps": 1280, "total_steps": 3322, "loss": 0.7281, "learning_rate": 0.00015442129132466054, "epoch": 0.3852520692249812, "percentage": 38.53, "elapsed_time": "0:53:08", "remaining_time": "1:24:46"}
+{"current_steps": 1290, "total_steps": 3322, "loss": 0.7471, "learning_rate": 0.00015353652503048384, "epoch": 0.38826185101580135, "percentage": 38.83, "elapsed_time": "0:53:32", "remaining_time": "1:24:19"}
+{"current_steps": 1300, "total_steps": 3322, "loss": 0.7469, "learning_rate": 0.00015264584456233502, "epoch": 0.3912716328066215, "percentage": 39.13, "elapsed_time": "0:53:55", "remaining_time": "1:23:52"}
+{"current_steps": 1310, "total_steps": 3322, "loss": 0.7945, "learning_rate": 0.0001517493483135864, "epoch": 0.3942814145974417, "percentage": 39.43, "elapsed_time": "0:54:18", "remaining_time": "1:23:25"}
+{"current_steps": 1320, "total_steps": 3322, "loss": 0.7659, "learning_rate": 0.00015084713532007905, "epoch": 0.3972911963882618, "percentage": 39.74, "elapsed_time": "0:54:42", "remaining_time": "1:22:57"}
+{"current_steps": 1330, "total_steps": 3322, "loss": 0.7718, "learning_rate": 0.00014993930524918208, "epoch": 0.400300978179082, "percentage": 40.04, "elapsed_time": "0:55:05", "remaining_time": "1:22:30"}
+{"current_steps": 1340, "total_steps": 3322, "loss": 0.759, "learning_rate": 0.00014902595838878256, "epoch": 0.4033107599699022, "percentage": 40.34, "elapsed_time": "0:55:28", "remaining_time": "1:22:03"}
+{"current_steps": 1350, "total_steps": 3322, "loss": 0.7568, "learning_rate": 0.0001481071956362067, "epoch": 0.40632054176072235, "percentage": 40.64, "elapsed_time": "0:55:52", "remaining_time": "1:21:37"}
+{"current_steps": 1360, "total_steps": 3322, "loss": 0.7504, "learning_rate": 0.0001471831184870737, "epoch": 0.40933032355154253, "percentage": 40.94, "elapsed_time": "0:56:16", "remaining_time": "1:21:11"}
+{"current_steps": 1370, "total_steps": 3322, "loss": 0.7365, "learning_rate": 0.00014625382902408356, "epoch": 0.4123401053423627, "percentage": 41.24, "elapsed_time": "0:56:40", "remaining_time": "1:20:44"}
+{"current_steps": 1380, "total_steps": 3322, "loss": 0.7444, "learning_rate": 0.00014531942990573998, "epoch": 0.4153498871331828, "percentage": 41.54, "elapsed_time": "0:57:04", "remaining_time": "1:20:18"}
+{"current_steps": 1390, "total_steps": 3322, "loss": 0.7574, "learning_rate": 0.00014438002435500979, "epoch": 0.418359668924003, "percentage": 41.84, "elapsed_time": "0:57:28", "remaining_time": "1:19:52"}
+{"current_steps": 1400, "total_steps": 3322, "loss": 0.74, "learning_rate": 0.0001434357161479198, "epoch": 0.4213694507148232, "percentage": 42.14, "elapsed_time": "0:57:51", "remaining_time": "1:19:26"}
+{"current_steps": 1410, "total_steps": 3322, "loss": 0.7761, "learning_rate": 0.0001424866096020927, "epoch": 0.42437923250564336, "percentage": 42.44, "elapsed_time": "0:58:13", "remaining_time": "1:18:57"}
+{"current_steps": 1420, "total_steps": 3322, "loss": 0.7895, "learning_rate": 0.00014153280956522322, "epoch": 0.42738901429646353, "percentage": 42.75, "elapsed_time": "0:58:36", "remaining_time": "1:18:29"}
+{"current_steps": 1430, "total_steps": 3322, "loss": 0.794, "learning_rate": 0.00014057442140349543, "epoch": 0.43039879608728365, "percentage": 43.05, "elapsed_time": "0:58:59", "remaining_time": "1:18:02"}
+{"current_steps": 1440, "total_steps": 3322, "loss": 0.7471, "learning_rate": 0.00013961155098994309, "epoch": 0.43340857787810383, "percentage": 43.35, "elapsed_time": "0:59:21", "remaining_time": "1:17:34"}
+{"current_steps": 1450, "total_steps": 3322, "loss": 0.745, "learning_rate": 0.00013864430469275377, "epoch": 0.436418359668924, "percentage": 43.65, "elapsed_time": "0:59:45", "remaining_time": "1:17:08"}
+{"current_steps": 1460, "total_steps": 3322, "loss": 0.7392, "learning_rate": 0.00013767278936351854, "epoch": 0.4394281414597442, "percentage": 43.95, "elapsed_time": "1:00:08", "remaining_time": "1:16:42"}
+{"current_steps": 1470, "total_steps": 3322, "loss": 0.7486, "learning_rate": 0.00013669711232542776, "epoch": 0.44243792325056436, "percentage": 44.25, "elapsed_time": "1:00:32", "remaining_time": "1:16:16"}
+{"current_steps": 1480, "total_steps": 3322, "loss": 0.7571, "learning_rate": 0.00013571738136141555, "epoch": 0.4454477050413845, "percentage": 44.55, "elapsed_time": "1:00:56", "remaining_time": "1:15:50"}
+{"current_steps": 1490, "total_steps": 3322, "loss": 0.76, "learning_rate": 0.0001347337047022526, "epoch": 0.44845748683220465, "percentage": 44.85, "elapsed_time": "1:01:19", "remaining_time": "1:15:24"}
+{"current_steps": 1500, "total_steps": 3322, "loss": 0.7444, "learning_rate": 0.00013374619101459012, "epoch": 0.45146726862302483, "percentage": 45.15, "elapsed_time": "1:01:42", "remaining_time": "1:14:57"}
+{"current_steps": 1510, "total_steps": 3322, "loss": 0.7755, "learning_rate": 0.00013275494938895556, "epoch": 0.454477050413845, "percentage": 45.45, "elapsed_time": "1:02:06", "remaining_time": "1:14:31"}
+{"current_steps": 1520, "total_steps": 3322, "loss": 0.755, "learning_rate": 0.00013176008932770113, "epoch": 0.4574868322046652, "percentage": 45.76, "elapsed_time": "1:02:30", "remaining_time": "1:14:06"}
+{"current_steps": 1530, "total_steps": 3322, "loss": 0.7369, "learning_rate": 0.00013076172073290724, "epoch": 0.4604966139954853, "percentage": 46.06, "elapsed_time": "1:02:53", "remaining_time": "1:13:40"}
+{"current_steps": 1540, "total_steps": 3322, "loss": 0.7773, "learning_rate": 0.00012975995389424166, "epoch": 0.4635063957863055, "percentage": 46.36, "elapsed_time": "1:03:17", "remaining_time": "1:13:14"}
+{"current_steps": 1550, "total_steps": 3322, "loss": 0.7287, "learning_rate": 0.0001287548994767758, "epoch": 0.46651617757712566, "percentage": 46.66, "elapsed_time": "1:03:42", "remaining_time": "1:12:49"}
+{"current_steps": 1560, "total_steps": 3322, "loss": 0.7717, "learning_rate": 0.00012774666850875942, "epoch": 0.46952595936794583, "percentage": 46.96, "elapsed_time": "1:04:05", "remaining_time": "1:12:23"}
+{"current_steps": 1570, "total_steps": 3322, "loss": 0.7613, "learning_rate": 0.00012673537236935556, "epoch": 0.472535741158766, "percentage": 47.26, "elapsed_time": "1:04:28", "remaining_time": "1:11:57"}
+{"current_steps": 1580, "total_steps": 3322, "loss": 0.7602, "learning_rate": 0.00012572112277633649, "epoch": 0.47554552294958613, "percentage": 47.56, "elapsed_time": "1:04:52", "remaining_time": "1:11:31"}
+{"current_steps": 1590, "total_steps": 3322, "loss": 0.7241, "learning_rate": 0.0001247040317737419, "epoch": 0.4785553047404063, "percentage": 47.86, "elapsed_time": "1:05:15", "remaining_time": "1:11:05"}
+{"current_steps": 1600, "total_steps": 3322, "loss": 0.7313, "learning_rate": 0.00012368421171950192, "epoch": 0.4815650865312265, "percentage": 48.16, "elapsed_time": "1:05:39", "remaining_time": "1:10:40"}
+{"current_steps": 1610, "total_steps": 3322, "loss": 0.7432, "learning_rate": 0.00012266177527302472, "epoch": 0.48457486832204666, "percentage": 48.46, "elapsed_time": "1:06:03", "remaining_time": "1:10:14"}
+{"current_steps": 1620, "total_steps": 3322, "loss": 0.7599, "learning_rate": 0.0001216368353827508, "epoch": 0.48758465011286684, "percentage": 48.77, "elapsed_time": "1:06:27", "remaining_time": "1:09:49"}
+{"current_steps": 1630, "total_steps": 3322, "loss": 0.7386, "learning_rate": 0.00012060950527367603, "epoch": 0.49059443190368696, "percentage": 49.07, "elapsed_time": "1:06:51", "remaining_time": "1:09:24"}
+{"current_steps": 1640, "total_steps": 3322, "loss": 0.7548, "learning_rate": 0.00011957989843484345, "epoch": 0.49360421369450713, "percentage": 49.37, "elapsed_time": "1:07:15", "remaining_time": "1:08:58"}
+{"current_steps": 1650, "total_steps": 3322, "loss": 0.7838, "learning_rate": 0.00011854812860680613, "epoch": 0.4966139954853273, "percentage": 49.67, "elapsed_time": "1:07:38", "remaining_time": "1:08:32"}
+{"current_steps": 1660, "total_steps": 3322, "loss": 0.7492, "learning_rate": 0.00011751430976906233, "epoch": 0.4996237772761475, "percentage": 49.97, "elapsed_time": "1:08:02", "remaining_time": "1:08:06"}
+{"current_steps": 1670, "total_steps": 3322, "loss": 0.7897, "learning_rate": 0.00011647855612746423, "epoch": 0.5026335590669676, "percentage": 50.27, "elapsed_time": "1:08:25", "remaining_time": "1:07:41"}
+{"current_steps": 1680, "total_steps": 3322, "loss": 0.7847, "learning_rate": 0.00011544098210160152, "epoch": 0.5056433408577878, "percentage": 50.57, "elapsed_time": "1:08:50", "remaining_time": "1:07:16"}
+{"current_steps": 1690, "total_steps": 3322, "loss": 0.7624, "learning_rate": 0.00011440170231216154, "epoch": 0.508653122648608, "percentage": 50.87, "elapsed_time": "1:09:12", "remaining_time": "1:06:49"}
+{"current_steps": 1700, "total_steps": 3322, "loss": 0.7973, "learning_rate": 0.00011336083156826722, "epoch": 0.5116629044394282, "percentage": 51.17, "elapsed_time": "1:09:35", "remaining_time": "1:06:24"}
+{"current_steps": 1710, "total_steps": 3322, "loss": 0.7322, "learning_rate": 0.00011231848485479395, "epoch": 0.5146726862302483, "percentage": 51.48, "elapsed_time": "1:10:00", "remaining_time": "1:05:59"}
+{"current_steps": 1720, "total_steps": 3322, "loss": 0.7449, "learning_rate": 0.00011127477731966735, "epoch": 0.5176824680210684, "percentage": 51.78, "elapsed_time": "1:10:23", "remaining_time": "1:05:33"}
+{"current_steps": 1730, "total_steps": 3322, "loss": 0.7988, "learning_rate": 0.00011022982426114292, "epoch": 0.5206922498118887, "percentage": 52.08, "elapsed_time": "1:10:46", "remaining_time": "1:05:07"}
+{"current_steps": 1740, "total_steps": 3322, "loss": 0.7273, "learning_rate": 0.00010918374111506893, "epoch": 0.5237020316027088, "percentage": 52.38, "elapsed_time": "1:11:10", "remaining_time": "1:04:42"}
+{"current_steps": 1750, "total_steps": 3322, "loss": 0.7367, "learning_rate": 0.00010813664344213427, "epoch": 0.526711813393529, "percentage": 52.68, "elapsed_time": "1:11:34", "remaining_time": "1:04:17"}
+{"current_steps": 1760, "total_steps": 3322, "loss": 0.7702, "learning_rate": 0.00010708864691510254, "epoch": 0.5297215951843491, "percentage": 52.98, "elapsed_time": "1:11:57", "remaining_time": "1:03:51"}
+{"current_steps": 1770, "total_steps": 3322, "loss": 0.7853, "learning_rate": 0.00010603986730603368, "epoch": 0.5327313769751693, "percentage": 53.28, "elapsed_time": "1:12:20", "remaining_time": "1:03:26"}
+{"current_steps": 1780, "total_steps": 3322, "loss": 0.7576, "learning_rate": 0.00010499042047349455, "epoch": 0.5357411587659895, "percentage": 53.58, "elapsed_time": "1:12:44", "remaining_time": "1:03:00"}
+{"current_steps": 1790, "total_steps": 3322, "loss": 0.7363, "learning_rate": 0.00010394042234976016, "epoch": 0.5387509405568096, "percentage": 53.88, "elapsed_time": "1:13:08", "remaining_time": "1:02:35"}
+{"current_steps": 1800, "total_steps": 3322, "loss": 0.7501, "learning_rate": 0.00010288998892800657, "epoch": 0.5417607223476298, "percentage": 54.18, "elapsed_time": "1:13:31", "remaining_time": "1:02:09"}
+{"current_steps": 1810, "total_steps": 3322, "loss": 0.7768, "learning_rate": 0.0001018392362494972, "epoch": 0.54477050413845, "percentage": 54.49, "elapsed_time": "1:13:54", "remaining_time": "1:01:44"}
+{"current_steps": 1820, "total_steps": 3322, "loss": 0.7803, "learning_rate": 0.00010078828039076367, "epoch": 0.5477802859292701, "percentage": 54.79, "elapsed_time": "1:14:17", "remaining_time": "1:01:18"}
+{"current_steps": 1830, "total_steps": 3322, "loss": 0.7686, "learning_rate": 9.973723745078296e-05, "epoch": 0.5507900677200903, "percentage": 55.09, "elapsed_time": "1:14:41", "remaining_time": "1:00:53"}
+{"current_steps": 1840, "total_steps": 3322, "loss": 0.7623, "learning_rate": 9.868622353815188e-05, "epoch": 0.5537998495109104, "percentage": 55.39, "elapsed_time": "1:15:04", "remaining_time": "1:00:28"}
+{"current_steps": 1850, "total_steps": 3322, "loss": 0.7439, "learning_rate": 9.763535475826054e-05, "epoch": 0.5568096313017307, "percentage": 55.69, "elapsed_time": "1:15:29", "remaining_time": "1:00:03"}
+{"current_steps": 1860, "total_steps": 3322, "loss": 0.7825, "learning_rate": 9.658474720046637e-05, "epoch": 0.5598194130925508, "percentage": 55.99, "elapsed_time": "1:15:52", "remaining_time": "0:59:38"}
+{"current_steps": 1870, "total_steps": 3322, "loss": 0.7802, "learning_rate": 9.553451692526954e-05, "epoch": 0.5628291948833709, "percentage": 56.29, "elapsed_time": "1:16:15", "remaining_time": "0:59:12"}
+{"current_steps": 1880, "total_steps": 3322, "loss": 0.7328, "learning_rate": 9.448477995149182e-05, "epoch": 0.5658389766741911, "percentage": 56.59, "elapsed_time": "1:16:39", "remaining_time": "0:58:47"}
+{"current_steps": 1890, "total_steps": 3322, "loss": 0.7464, "learning_rate": 9.343565224346013e-05, "epoch": 0.5688487584650113, "percentage": 56.89, "elapsed_time": "1:17:02", "remaining_time": "0:58:22"}
+{"current_steps": 1900, "total_steps": 3322, "loss": 0.7387, "learning_rate": 9.238724969819579e-05, "epoch": 0.5718585402558315, "percentage": 57.19, "elapsed_time": "1:17:25", "remaining_time": "0:57:57"}
+{"current_steps": 1910, "total_steps": 3322, "loss": 0.7136, "learning_rate": 9.13396881326115e-05, "epoch": 0.5748683220466516, "percentage": 57.5, "elapsed_time": "1:17:50", "remaining_time": "0:57:32"}
+{"current_steps": 1920, "total_steps": 3322, "loss": 0.7702, "learning_rate": 9.029308327071702e-05, "epoch": 0.5778781038374717, "percentage": 57.8, "elapsed_time": "1:18:13", "remaining_time": "0:57:07"}
+{"current_steps": 1930, "total_steps": 3322, "loss": 0.7901, "learning_rate": 8.924755073083517e-05, "epoch": 0.580887885628292, "percentage": 58.1, "elapsed_time": "1:18:37", "remaining_time": "0:56:42"}
+{"current_steps": 1940, "total_steps": 3322, "loss": 0.7771, "learning_rate": 8.820320601282949e-05, "epoch": 0.5838976674191121, "percentage": 58.4, "elapsed_time": "1:19:00", "remaining_time": "0:56:16"}
+{"current_steps": 1950, "total_steps": 3322, "loss": 0.7601, "learning_rate": 8.71601644853449e-05, "epoch": 0.5869074492099323, "percentage": 58.7, "elapsed_time": "1:19:23", "remaining_time": "0:55:51"}
+{"current_steps": 1960, "total_steps": 3322, "loss": 0.7575, "learning_rate": 8.61185413730631e-05, "epoch": 0.5899172310007524, "percentage": 59.0, "elapsed_time": "1:19:45", "remaining_time": "0:55:25"}
+{"current_steps": 1970, "total_steps": 3322, "loss": 0.8136, "learning_rate": 8.507845174397357e-05, "epoch": 0.5929270127915726, "percentage": 59.3, "elapsed_time": "1:20:08", "remaining_time": "0:54:59"}
+{"current_steps": 1980, "total_steps": 3322, "loss": 0.7205, "learning_rate": 8.404001049666211e-05, "epoch": 0.5959367945823928, "percentage": 59.6, "elapsed_time": "1:20:32", "remaining_time": "0:54:35"}
+{"current_steps": 1990, "total_steps": 3322, "loss": 0.768, "learning_rate": 8.300333234761787e-05, "epoch": 0.5989465763732129, "percentage": 59.9, "elapsed_time": "1:20:55", "remaining_time": "0:54:09"}
+{"current_steps": 2000, "total_steps": 3322, "loss": 0.7502, "learning_rate": 8.196853181856081e-05, "epoch": 0.6019563581640331, "percentage": 60.2, "elapsed_time": "1:21:18", "remaining_time": "0:53:44"}
+{"current_steps": 2000, "total_steps": 3322, "eval_loss": 0.8055371046066284, "epoch": 0.6019563581640331, "percentage": 60.2, "elapsed_time": "1:23:41", "remaining_time": "0:55:19"}
+{"current_steps": 2010, "total_steps": 3322, "loss": 0.7597, "learning_rate": 8.093572322379045e-05, "epoch": 0.6049661399548533, "percentage": 60.51, "elapsed_time": "1:24:06", "remaining_time": "0:54:54"}
+{"current_steps": 2020, "total_steps": 3322, "loss": 0.7592, "learning_rate": 7.990502065755748e-05, "epoch": 0.6079759217456734, "percentage": 60.81, "elapsed_time": "1:24:30", "remaining_time": "0:54:28"}
+{"current_steps": 2030, "total_steps": 3322, "loss": 0.7432, "learning_rate": 7.887653798145987e-05, "epoch": 0.6109857035364936, "percentage": 61.11, "elapsed_time": "1:24:54", "remaining_time": "0:54:02"}
+{"current_steps": 2040, "total_steps": 3322, "loss": 0.7638, "learning_rate": 7.785038881186462e-05, "epoch": 0.6139954853273137, "percentage": 61.41, "elapsed_time": "1:25:17", "remaining_time": "0:53:35"}
+{"current_steps": 2050, "total_steps": 3322, "loss": 0.7451, "learning_rate": 7.682668650735645e-05, "epoch": 0.617005267118134, "percentage": 61.71, "elapsed_time": "1:25:41", "remaining_time": "0:53:10"}
+{"current_steps": 2060, "total_steps": 3322, "loss": 0.7432, "learning_rate": 7.580554415621522e-05, "epoch": 0.6200150489089541, "percentage": 62.01, "elapsed_time": "1:26:04", "remaining_time": "0:52:43"}
+{"current_steps": 2070, "total_steps": 3322, "loss": 0.7924, "learning_rate": 7.478707456392302e-05, "epoch": 0.6230248306997742, "percentage": 62.31, "elapsed_time": "1:26:27", "remaining_time": "0:52:17"}
+{"current_steps": 2080, "total_steps": 3322, "loss": 0.7471, "learning_rate": 7.377139024070254e-05, "epoch": 0.6260346124905944, "percentage": 62.61, "elapsed_time": "1:26:52", "remaining_time": "0:51:52"}
+{"current_steps": 2090, "total_steps": 3322, "loss": 0.7535, "learning_rate": 7.275860338908815e-05, "epoch": 0.6290443942814146, "percentage": 62.91, "elapsed_time": "1:27:15", "remaining_time": "0:51:26"}
+{"current_steps": 2100, "total_steps": 3322, "loss": 0.7523, "learning_rate": 7.174882589153076e-05, "epoch": 0.6320541760722348, "percentage": 63.21, "elapsed_time": "1:27:39", "remaining_time": "0:51:00"}
+{"current_steps": 2110, "total_steps": 3322, "loss": 0.7507, "learning_rate": 7.07421692980384e-05, "epoch": 0.6350639578630549, "percentage": 63.52, "elapsed_time": "1:28:03", "remaining_time": "0:50:34"}
+{"current_steps": 2120, "total_steps": 3322, "loss": 0.7671, "learning_rate": 6.973874481385312e-05, "epoch": 0.6380737396538751, "percentage": 63.82, "elapsed_time": "1:28:27", "remaining_time": "0:50:08"}
+{"current_steps": 2130, "total_steps": 3322, "loss": 0.7595, "learning_rate": 6.873866328716614e-05, "epoch": 0.6410835214446953, "percentage": 64.12, "elapsed_time": "1:28:50", "remaining_time": "0:49:42"}
+{"current_steps": 2140, "total_steps": 3322, "loss": 0.7384, "learning_rate": 6.774203519687265e-05, "epoch": 0.6440933032355154, "percentage": 64.42, "elapsed_time": "1:29:13", "remaining_time": "0:49:17"}
+{"current_steps": 2150, "total_steps": 3322, "loss": 0.7629, "learning_rate": 6.674897064036706e-05, "epoch": 0.6471030850263356, "percentage": 64.72, "elapsed_time": "1:29:36", "remaining_time": "0:48:50"}
+{"current_steps": 2160, "total_steps": 3322, "loss": 0.7793, "learning_rate": 6.575957932138057e-05, "epoch": 0.6501128668171557, "percentage": 65.02, "elapsed_time": "1:29:59", "remaining_time": "0:48:24"}
+{"current_steps": 2170, "total_steps": 3322, "loss": 0.7586, "learning_rate": 6.47739705378623e-05, "epoch": 0.653122648607976, "percentage": 65.32, "elapsed_time": "1:30:23", "remaining_time": "0:47:58"}
+{"current_steps": 2180, "total_steps": 3322, "loss": 0.7564, "learning_rate": 6.379225316990505e-05, "epoch": 0.6561324303987961, "percentage": 65.62, "elapsed_time": "1:30:47", "remaining_time": "0:47:33"}
+{"current_steps": 2190, "total_steps": 3322, "loss": 0.7354, "learning_rate": 6.281453566771735e-05, "epoch": 0.6591422121896162, "percentage": 65.92, "elapsed_time": "1:31:11", "remaining_time": "0:47:08"}
+{"current_steps": 2200, "total_steps": 3322, "loss": 0.8203, "learning_rate": 6.184092603964308e-05, "epoch": 0.6621519939804364, "percentage": 66.23, "elapsed_time": "1:31:34", "remaining_time": "0:46:42"}
+{"current_steps": 2210, "total_steps": 3322, "loss": 0.7466, "learning_rate": 6.087153184022969e-05, "epoch": 0.6651617757712566, "percentage": 66.53, "elapsed_time": "1:31:57", "remaining_time": "0:46:16"}
+{"current_steps": 2220, "total_steps": 3322, "loss": 0.76, "learning_rate": 5.990646015834668e-05, "epoch": 0.6681715575620768, "percentage": 66.83, "elapsed_time": "1:32:21", "remaining_time": "0:45:50"}
+{"current_steps": 2230, "total_steps": 3322, "loss": 0.7662, "learning_rate": 5.894581760535549e-05, "epoch": 0.6711813393528969, "percentage": 67.13, "elapsed_time": "1:32:44", "remaining_time": "0:45:24"}
+{"current_steps": 2240, "total_steps": 3322, "loss": 0.7742, "learning_rate": 5.798971030333227e-05, "epoch": 0.674191121143717, "percentage": 67.43, "elapsed_time": "1:33:08", "remaining_time": "0:44:59"}
+{"current_steps": 2250, "total_steps": 3322, "loss": 0.7837, "learning_rate": 5.703824387334442e-05, "epoch": 0.6772009029345373, "percentage": 67.73, "elapsed_time": "1:33:34", "remaining_time": "0:44:34"}
+{"current_steps": 2260, "total_steps": 3322, "loss": 0.758, "learning_rate": 5.609152342378278e-05, "epoch": 0.6802106847253574, "percentage": 68.03, "elapsed_time": "1:33:56", "remaining_time": "0:44:08"}
+{"current_steps": 2270, "total_steps": 3322, "loss": 0.726, "learning_rate": 5.514965353875019e-05, "epoch": 0.6832204665161776, "percentage": 68.33, "elapsed_time": "1:34:20", "remaining_time": "0:43:43"}
+{"current_steps": 2280, "total_steps": 3322, "loss": 0.7199, "learning_rate": 5.4212738266508245e-05, "epoch": 0.6862302483069977, "percentage": 68.63, "elapsed_time": "1:34:44", "remaining_time": "0:43:17"}
+{"current_steps": 2290, "total_steps": 3322, "loss": 0.7379, "learning_rate": 5.3280881107982946e-05, "epoch": 0.6892400300978179, "percentage": 68.93, "elapsed_time": "1:35:07", "remaining_time": "0:42:52"}
+{"current_steps": 2300, "total_steps": 3322, "loss": 0.7574, "learning_rate": 5.235418500533109e-05, "epoch": 0.6922498118886381, "percentage": 69.24, "elapsed_time": "1:35:31", "remaining_time": "0:42:26"}
+{"current_steps": 2310, "total_steps": 3322, "loss": 0.7737, "learning_rate": 5.143275233056817e-05, "epoch": 0.6952595936794582, "percentage": 69.54, "elapsed_time": "1:35:54", "remaining_time": "0:42:00"}
+{"current_steps": 2320, "total_steps": 3322, "loss": 0.7268, "learning_rate": 5.051668487425938e-05, "epoch": 0.6982693754702785, "percentage": 69.84, "elapsed_time": "1:36:18", "remaining_time": "0:41:35"}
+{"current_steps": 2330, "total_steps": 3322, "loss": 0.7542, "learning_rate": 4.960608383427481e-05, "epoch": 0.7012791572610986, "percentage": 70.14, "elapsed_time": "1:36:41", "remaining_time": "0:41:10"}
+{"current_steps": 2340, "total_steps": 3322, "loss": 0.7493, "learning_rate": 4.8701049804610265e-05, "epoch": 0.7042889390519187, "percentage": 70.44, "elapsed_time": "1:37:04", "remaining_time": "0:40:44"}
+{"current_steps": 2350, "total_steps": 3322, "loss": 0.7468, "learning_rate": 4.780168276427441e-05, "epoch": 0.7072987208427389, "percentage": 70.74, "elapsed_time": "1:37:28", "remaining_time": "0:40:18"}
+{"current_steps": 2360, "total_steps": 3322, "loss": 0.7667, "learning_rate": 4.6908082066244275e-05, "epoch": 0.710308502633559, "percentage": 71.04, "elapsed_time": "1:37:51", "remaining_time": "0:39:53"}
+{"current_steps": 2370, "total_steps": 3322, "loss": 0.788, "learning_rate": 4.602034642648968e-05, "epoch": 0.7133182844243793, "percentage": 71.34, "elapsed_time": "1:38:14", "remaining_time": "0:39:27"}
+{"current_steps": 2380, "total_steps": 3322, "loss": 0.772, "learning_rate": 4.513857391306812e-05, "epoch": 0.7163280662151994, "percentage": 71.64, "elapsed_time": "1:38:38", "remaining_time": "0:39:02"}
+{"current_steps": 2390, "total_steps": 3322, "loss": 0.7257, "learning_rate": 4.4262861935291144e-05, "epoch": 0.7193378480060195, "percentage": 71.94, "elapsed_time": "1:39:02", "remaining_time": "0:38:37"}
+{"current_steps": 2400, "total_steps": 3322, "loss": 0.7456, "learning_rate": 4.339330723296373e-05, "epoch": 0.7223476297968398, "percentage": 72.25, "elapsed_time": "1:39:26", "remaining_time": "0:38:11"}
+{"current_steps": 2410, "total_steps": 3322, "loss": 0.7722, "learning_rate": 4.25300058656972e-05, "epoch": 0.7253574115876599, "percentage": 72.55, "elapsed_time": "1:39:49", "remaining_time": "0:37:46"}
+{"current_steps": 2420, "total_steps": 3322, "loss": 0.7871, "learning_rate": 4.1673053202297676e-05, "epoch": 0.7283671933784801, "percentage": 72.85, "elapsed_time": "1:40:12", "remaining_time": "0:37:21"}
+{"current_steps": 2430, "total_steps": 3322, "loss": 0.7394, "learning_rate": 4.0822543910230674e-05, "epoch": 0.7313769751693002, "percentage": 73.15, "elapsed_time": "1:40:36", "remaining_time": "0:36:55"}
+{"current_steps": 2440, "total_steps": 3322, "loss": 0.7318, "learning_rate": 3.997857194516319e-05, "epoch": 0.7343867569601203, "percentage": 73.45, "elapsed_time": "1:41:00", "remaining_time": "0:36:30"}
+{"current_steps": 2450, "total_steps": 3322, "loss": 0.7711, "learning_rate": 3.914123054058446e-05, "epoch": 0.7373965387509406, "percentage": 73.75, "elapsed_time": "1:41:24", "remaining_time": "0:36:05"}
+{"current_steps": 2460, "total_steps": 3322, "loss": 0.7341, "learning_rate": 3.831061219750636e-05, "epoch": 0.7404063205417607, "percentage": 74.05, "elapsed_time": "1:41:49", "remaining_time": "0:35:40"}
+{"current_steps": 2470, "total_steps": 3322, "loss": 0.7349, "learning_rate": 3.7486808674245047e-05, "epoch": 0.7434161023325809, "percentage": 74.35, "elapsed_time": "1:42:13", "remaining_time": "0:35:15"}
+{"current_steps": 2480, "total_steps": 3322, "loss": 0.7551, "learning_rate": 3.666991097628416e-05, "epoch": 0.746425884123401, "percentage": 74.65, "elapsed_time": "1:42:36", "remaining_time": "0:34:50"}
+{"current_steps": 2490, "total_steps": 3322, "loss": 0.7485, "learning_rate": 3.586000934622166e-05, "epoch": 0.7494356659142212, "percentage": 74.95, "elapsed_time": "1:42:59", "remaining_time": "0:34:24"}
+{"current_steps": 2500, "total_steps": 3322, "loss": 0.7308, "learning_rate": 3.5057193253800624e-05, "epoch": 0.7524454477050414, "percentage": 75.26, "elapsed_time": "1:43:22", "remaining_time": "0:33:59"}
+{"current_steps": 2510, "total_steps": 3322, "loss": 0.7717, "learning_rate": 3.426155138602558e-05, "epoch": 0.7554552294958615, "percentage": 75.56, "elapsed_time": "1:43:46", "remaining_time": "0:33:34"}
+{"current_steps": 2520, "total_steps": 3322, "loss": 0.753, "learning_rate": 3.347317163736524e-05, "epoch": 0.7584650112866818, "percentage": 75.86, "elapsed_time": "1:44:10", "remaining_time": "0:33:09"}
+{"current_steps": 2530, "total_steps": 3322, "loss": 0.7552, "learning_rate": 3.269214110004293e-05, "epoch": 0.7614747930775019, "percentage": 76.16, "elapsed_time": "1:44:34", "remaining_time": "0:32:44"}
+{"current_steps": 2540, "total_steps": 3322, "loss": 0.7146, "learning_rate": 3.191854605441527e-05, "epoch": 0.764484574868322, "percentage": 76.46, "elapsed_time": "1:44:58", "remaining_time": "0:32:19"}
+{"current_steps": 2550, "total_steps": 3322, "loss": 0.7733, "learning_rate": 3.115247195944102e-05, "epoch": 0.7674943566591422, "percentage": 76.76, "elapsed_time": "1:45:21", "remaining_time": "0:31:53"}
+{"current_steps": 2560, "total_steps": 3322, "loss": 0.7948, "learning_rate": 3.039400344324035e-05, "epoch": 0.7705041384499624, "percentage": 77.06, "elapsed_time": "1:45:44", "remaining_time": "0:31:28"}
+{"current_steps": 2570, "total_steps": 3322, "loss": 0.7637, "learning_rate": 2.9643224293745954e-05, "epoch": 0.7735139202407826, "percentage": 77.36, "elapsed_time": "1:46:07", "remaining_time": "0:31:03"}
+{"current_steps": 2580, "total_steps": 3322, "loss": 0.7326, "learning_rate": 2.8900217449447074e-05, "epoch": 0.7765237020316027, "percentage": 77.66, "elapsed_time": "1:46:30", "remaining_time": "0:30:37"}
+{"current_steps": 2590, "total_steps": 3322, "loss": 0.7571, "learning_rate": 2.8165064990227252e-05, "epoch": 0.7795334838224228, "percentage": 77.97, "elapsed_time": "1:46:54", "remaining_time": "0:30:12"}
+{"current_steps": 2600, "total_steps": 3322, "loss": 0.75, "learning_rate": 2.7437848128296982e-05, "epoch": 0.782543265613243, "percentage": 78.27, "elapsed_time": "1:47:17", "remaining_time": "0:29:47"}
+{"current_steps": 2610, "total_steps": 3322, "loss": 0.7693, "learning_rate": 2.6718647199222214e-05, "epoch": 0.7855530474040632, "percentage": 78.57, "elapsed_time": "1:47:41", "remaining_time": "0:29:22"}
+{"current_steps": 2620, "total_steps": 3322, "loss": 0.7499, "learning_rate": 2.600754165304966e-05, "epoch": 0.7885628291948834, "percentage": 78.87, "elapsed_time": "1:48:05", "remaining_time": "0:28:57"}
+{"current_steps": 2630, "total_steps": 3322, "loss": 0.7565, "learning_rate": 2.530461004553001e-05, "epoch": 0.7915726109857035, "percentage": 79.17, "elapsed_time": "1:48:28", "remaining_time": "0:28:32"}
+{"current_steps": 2640, "total_steps": 3322, "loss": 0.738, "learning_rate": 2.460993002943983e-05, "epoch": 0.7945823927765236, "percentage": 79.47, "elapsed_time": "1:48:53", "remaining_time": "0:28:07"}
+{"current_steps": 2650, "total_steps": 3322, "loss": 0.746, "learning_rate": 2.3923578346003363e-05, "epoch": 0.7975921745673439, "percentage": 79.77, "elapsed_time": "1:49:16", "remaining_time": "0:27:42"}
+{"current_steps": 2660, "total_steps": 3322, "loss": 0.7615, "learning_rate": 2.32456308164148e-05, "epoch": 0.800601956358164, "percentage": 80.07, "elapsed_time": "1:49:40", "remaining_time": "0:27:17"}
+{"current_steps": 2670, "total_steps": 3322, "loss": 0.8101, "learning_rate": 2.2576162333462402e-05, "epoch": 0.8036117381489842, "percentage": 80.37, "elapsed_time": "1:50:03", "remaining_time": "0:26:52"}
+{"current_steps": 2680, "total_steps": 3322, "loss": 0.7551, "learning_rate": 2.191524685325512e-05, "epoch": 0.8066215199398044, "percentage": 80.67, "elapsed_time": "1:50:26", "remaining_time": "0:26:27"}
+{"current_steps": 2690, "total_steps": 3322, "loss": 0.7621, "learning_rate": 2.126295738705262e-05, "epoch": 0.8096313017306245, "percentage": 80.98, "elapsed_time": "1:50:50", "remaining_time": "0:26:02"}
+{"current_steps": 2700, "total_steps": 3322, "loss": 0.7789, "learning_rate": 2.0619365993199747e-05, "epoch": 0.8126410835214447, "percentage": 81.28, "elapsed_time": "1:51:13", "remaining_time": "0:25:37"}
+{"current_steps": 2710, "total_steps": 3322, "loss": 0.7246, "learning_rate": 1.9984543769166265e-05, "epoch": 0.8156508653122648, "percentage": 81.58, "elapsed_time": "1:51:37", "remaining_time": "0:25:12"}
+{"current_steps": 2720, "total_steps": 3322, "loss": 0.7295, "learning_rate": 1.9358560843692787e-05, "epoch": 0.8186606471030851, "percentage": 81.88, "elapsed_time": "1:52:01", "remaining_time": "0:24:47"}
+{"current_steps": 2730, "total_steps": 3322, "loss": 0.7851, "learning_rate": 1.8741486369043505e-05, "epoch": 0.8216704288939052, "percentage": 82.18, "elapsed_time": "1:52:24", "remaining_time": "0:24:22"}
+{"current_steps": 2740, "total_steps": 3322, "loss": 0.7278, "learning_rate": 1.8133388513367078e-05, "epoch": 0.8246802106847254, "percentage": 82.48, "elapsed_time": "1:52:48", "remaining_time": "0:23:57"}
+{"current_steps": 2750, "total_steps": 3322, "loss": 0.7247, "learning_rate": 1.7534334453166068e-05, "epoch": 0.8276899924755455, "percentage": 82.78, "elapsed_time": "1:53:11", "remaining_time": "0:23:32"}
+{"current_steps": 2760, "total_steps": 3322, "loss": 0.7741, "learning_rate": 1.6944390365875952e-05, "epoch": 0.8306997742663657, "percentage": 83.08, "elapsed_time": "1:53:34", "remaining_time": "0:23:07"}
+{"current_steps": 2770, "total_steps": 3322, "loss": 0.7267, "learning_rate": 1.6363621422554476e-05, "epoch": 0.8337095560571859, "percentage": 83.38, "elapsed_time": "1:53:58", "remaining_time": "0:22:42"}
+{"current_steps": 2780, "total_steps": 3322, "loss": 0.7566, "learning_rate": 1.579209178068234e-05, "epoch": 0.836719337848006, "percentage": 83.68, "elapsed_time": "1:54:21", "remaining_time": "0:22:17"}
+{"current_steps": 2790, "total_steps": 3322, "loss": 0.8005, "learning_rate": 1.5229864577075547e-05, "epoch": 0.8397291196388262, "percentage": 83.99, "elapsed_time": "1:54:45", "remaining_time": "0:21:52"}
+{"current_steps": 2800, "total_steps": 3322, "loss": 0.7494, "learning_rate": 1.4677001920910827e-05, "epoch": 0.8427389014296464, "percentage": 84.29, "elapsed_time": "1:55:09", "remaining_time": "0:21:28"}
+{"current_steps": 2810, "total_steps": 3322, "loss": 0.7289, "learning_rate": 1.4133564886864381e-05, "epoch": 0.8457486832204665, "percentage": 84.59, "elapsed_time": "1:55:32", "remaining_time": "0:21:03"}
+{"current_steps": 2820, "total_steps": 3322, "loss": 0.7493, "learning_rate": 1.3599613508364984e-05, "epoch": 0.8487584650112867, "percentage": 84.89, "elapsed_time": "1:55:56", "remaining_time": "0:20:38"}
+{"current_steps": 2830, "total_steps": 3322, "loss": 0.7504, "learning_rate": 1.307520677096209e-05, "epoch": 0.8517682468021068, "percentage": 85.19, "elapsed_time": "1:56:19", "remaining_time": "0:20:13"}
+{"current_steps": 2840, "total_steps": 3322, "loss": 0.7705, "learning_rate": 1.2560402605809707e-05, "epoch": 0.8547780285929271, "percentage": 85.49, "elapsed_time": "1:56:44", "remaining_time": "0:19:48"}
+{"current_steps": 2850, "total_steps": 3322, "loss": 0.7307, "learning_rate": 1.2055257883266791e-05, "epoch": 0.8577878103837472, "percentage": 85.79, "elapsed_time": "1:57:06", "remaining_time": "0:19:23"}
+{"current_steps": 2860, "total_steps": 3322, "loss": 0.7741, "learning_rate": 1.1559828406614714e-05, "epoch": 0.8607975921745673, "percentage": 86.09, "elapsed_time": "1:57:29", "remaining_time": "0:18:58"}
+{"current_steps": 2870, "total_steps": 3322, "loss": 0.7238, "learning_rate": 1.1074168905892702e-05, "epoch": 0.8638073739653875, "percentage": 86.39, "elapsed_time": "1:57:52", "remaining_time": "0:18:33"}
+{"current_steps": 2880, "total_steps": 3322, "loss": 0.7315, "learning_rate": 1.0598333031851881e-05, "epoch": 0.8668171557562077, "percentage": 86.69, "elapsed_time": "1:58:16", "remaining_time": "0:18:09"}
+{"current_steps": 2890, "total_steps": 3322, "loss": 0.751, "learning_rate": 1.0132373350028313e-05, "epoch": 0.8698269375470279, "percentage": 87.0, "elapsed_time": "1:58:40", "remaining_time": "0:17:44"}
+{"current_steps": 2900, "total_steps": 3322, "loss": 0.72, "learning_rate": 9.676341334936346e-06, "epoch": 0.872836719337848, "percentage": 87.3, "elapsed_time": "1:59:04", "remaining_time": "0:17:19"}
+{"current_steps": 2910, "total_steps": 3322, "loss": 0.7433, "learning_rate": 9.230287364382007e-06, "epoch": 0.8758465011286681, "percentage": 87.6, "elapsed_time": "1:59:28", "remaining_time": "0:16:54"}
+{"current_steps": 2920, "total_steps": 3322, "loss": 0.7473, "learning_rate": 8.794260713897862e-06, "epoch": 0.8788562829194884, "percentage": 87.9, "elapsed_time": "1:59:51", "remaining_time": "0:16:29"}
+{"current_steps": 2930, "total_steps": 3322, "loss": 0.746, "learning_rate": 8.368309551299536e-06, "epoch": 0.8818660647103085, "percentage": 88.2, "elapsed_time": "2:00:13", "remaining_time": "0:16:05"}
+{"current_steps": 2940, "total_steps": 3322, "loss": 0.7523, "learning_rate": 7.952480931364658e-06, "epoch": 0.8848758465011287, "percentage": 88.5, "elapsed_time": "2:00:38", "remaining_time": "0:15:40"}
+{"current_steps": 2950, "total_steps": 3322, "loss": 0.7359, "learning_rate": 7.546820790634646e-06, "epoch": 0.8878856282919488, "percentage": 88.8, "elapsed_time": "2:01:02", "remaining_time": "0:15:15"}
+{"current_steps": 2960, "total_steps": 3322, "loss": 0.7201, "learning_rate": 7.1513739423402e-06, "epoch": 0.890895410082769, "percentage": 89.1, "elapsed_time": "2:01:26", "remaining_time": "0:14:51"}
+{"current_steps": 2970, "total_steps": 3322, "loss": 0.7619, "learning_rate": 6.766184071450721e-06, "epoch": 0.8939051918735892, "percentage": 89.4, "elapsed_time": "2:01:49", "remaining_time": "0:14:26"}
+{"current_steps": 2980, "total_steps": 3322, "loss": 0.7654, "learning_rate": 6.391293729848435e-06, "epoch": 0.8969149736644093, "percentage": 89.7, "elapsed_time": "2:02:13", "remaining_time": "0:14:01"}
+{"current_steps": 2990, "total_steps": 3322, "loss": 0.7416, "learning_rate": 6.026744331627731e-06, "epoch": 0.8999247554552295, "percentage": 90.01, "elapsed_time": "2:02:36", "remaining_time": "0:13:36"}
+{"current_steps": 3000, "total_steps": 3322, "loss": 0.7516, "learning_rate": 5.672576148520137e-06, "epoch": 0.9029345372460497, "percentage": 90.31, "elapsed_time": "2:03:00", "remaining_time": "0:13:12"}
+{"current_steps": 3000, "total_steps": 3322, "eval_loss": 0.7987983226776123, "epoch": 0.9029345372460497, "percentage": 90.31, "elapsed_time": "2:05:22", "remaining_time": "0:13:27"}
+{"current_steps": 3010, "total_steps": 3322, "loss": 0.7357, "learning_rate": 5.328828305445477e-06, "epoch": 0.9059443190368698, "percentage": 90.61, "elapsed_time": "2:05:47", "remaining_time": "0:13:02"}
+{"current_steps": 3020, "total_steps": 3322, "loss": 0.7725, "learning_rate": 4.9955387761897785e-06, "epoch": 0.90895410082769, "percentage": 90.91, "elapsed_time": "2:06:10", "remaining_time": "0:12:37"}
+{"current_steps": 3030, "total_steps": 3322, "loss": 0.776, "learning_rate": 4.672744379210336e-06, "epoch": 0.9119638826185101, "percentage": 91.21, "elapsed_time": "2:06:34", "remaining_time": "0:12:11"}
+{"current_steps": 3040, "total_steps": 3322, "loss": 0.7323, "learning_rate": 4.360480773568321e-06, "epoch": 0.9149736644093304, "percentage": 91.51, "elapsed_time": "2:06:58", "remaining_time": "0:11:46"}
+{"current_steps": 3050, "total_steps": 3322, "loss": 0.7436, "learning_rate": 4.058782454989529e-06, "epoch": 0.9179834462001505, "percentage": 91.81, "elapsed_time": "2:07:21", "remaining_time": "0:11:21"}
+{"current_steps": 3060, "total_steps": 3322, "loss": 0.7538, "learning_rate": 3.767682752053714e-06, "epoch": 0.9209932279909706, "percentage": 92.11, "elapsed_time": "2:07:45", "remaining_time": "0:10:56"}
+{"current_steps": 3070, "total_steps": 3322, "loss": 0.7535, "learning_rate": 3.487213822512714e-06, "epoch": 0.9240030097817908, "percentage": 92.41, "elapsed_time": "2:08:08", "remaining_time": "0:10:31"}
+{"current_steps": 3080, "total_steps": 3322, "loss": 0.7572, "learning_rate": 3.2174066497380086e-06, "epoch": 0.927012791572611, "percentage": 92.72, "elapsed_time": "2:08:32", "remaining_time": "0:10:06"}
+{"current_steps": 3090, "total_steps": 3322, "loss": 0.7114, "learning_rate": 2.958291039298e-06, "epoch": 0.9300225733634312, "percentage": 93.02, "elapsed_time": "2:08:56", "remaining_time": "0:09:40"}
+{"current_steps": 3100, "total_steps": 3322, "loss": 0.7656, "learning_rate": 2.709895615665392e-06, "epoch": 0.9330323551542513, "percentage": 93.32, "elapsed_time": "2:09:20", "remaining_time": "0:09:15"}
+{"current_steps": 3110, "total_steps": 3322, "loss": 0.7361, "learning_rate": 2.472247819055029e-06, "epoch": 0.9360421369450714, "percentage": 93.62, "elapsed_time": "2:09:44", "remaining_time": "0:08:50"}
+{"current_steps": 3120, "total_steps": 3322, "loss": 0.7789, "learning_rate": 2.2453739023926113e-06, "epoch": 0.9390519187358917, "percentage": 93.92, "elapsed_time": "2:10:08", "remaining_time": "0:08:25"}
+{"current_steps": 3130, "total_steps": 3322, "loss": 0.7756, "learning_rate": 2.0292989284144915e-06, "epoch": 0.9420617005267118, "percentage": 94.22, "elapsed_time": "2:10:32", "remaining_time": "0:08:00"}
+{"current_steps": 3140, "total_steps": 3322, "loss": 0.7319, "learning_rate": 1.8240467668990457e-06, "epoch": 0.945071482317532, "percentage": 94.52, "elapsed_time": "2:10:57", "remaining_time": "0:07:35"}
+{"current_steps": 3150, "total_steps": 3322, "loss": 0.7469, "learning_rate": 1.6296400920297383e-06, "epoch": 0.9480812641083521, "percentage": 94.82, "elapsed_time": "2:11:21", "remaining_time": "0:07:10"}
+{"current_steps": 3160, "total_steps": 3322, "loss": 0.7587, "learning_rate": 1.4461003798903695e-06, "epoch": 0.9510910458991723, "percentage": 95.12, "elapsed_time": "2:11:44", "remaining_time": "0:06:45"}
+{"current_steps": 3170, "total_steps": 3322, "loss": 0.7504, "learning_rate": 1.2734479060925397e-06, "epoch": 0.9541008276899925, "percentage": 95.42, "elapsed_time": "2:12:08", "remaining_time": "0:06:20"}
+{"current_steps": 3180, "total_steps": 3322, "loss": 0.7628, "learning_rate": 1.1117017435358423e-06, "epoch": 0.9571106094808126, "percentage": 95.73, "elapsed_time": "2:12:31", "remaining_time": "0:05:55"}
+{"current_steps": 3190, "total_steps": 3322, "loss": 0.7195, "learning_rate": 9.608797603008812e-07, "epoch": 0.9601203912716328, "percentage": 96.03, "elapsed_time": "2:12:55", "remaining_time": "0:05:30"}
+{"current_steps": 3200, "total_steps": 3322, "loss": 0.7476, "learning_rate": 8.209986176753948e-07, "epoch": 0.963130173062453, "percentage": 96.33, "elapsed_time": "2:13:19", "remaining_time": "0:05:04"}
+{"current_steps": 3210, "total_steps": 3322, "loss": 0.7578, "learning_rate": 6.920737683136613e-07, "epoch": 0.9661399548532731, "percentage": 96.63, "elapsed_time": "2:13:42", "remaining_time": "0:04:39"}
+{"current_steps": 3220, "total_steps": 3322, "loss": 0.7537, "learning_rate": 5.741194545294648e-07, "epoch": 0.9691497366440933, "percentage": 96.93, "elapsed_time": "2:14:07", "remaining_time": "0:04:14"}
+{"current_steps": 3230, "total_steps": 3322, "loss": 0.7226, "learning_rate": 4.671487067227531e-07, "epoch": 0.9721595184349134, "percentage": 97.23, "elapsed_time": "2:14:30", "remaining_time": "0:03:49"}
+{"current_steps": 3240, "total_steps": 3322, "loss": 0.7487, "learning_rate": 3.711733419401453e-07, "epoch": 0.9751693002257337, "percentage": 97.53, "elapsed_time": "2:14:53", "remaining_time": "0:03:24"}
+{"current_steps": 3250, "total_steps": 3322, "loss": 0.773, "learning_rate": 2.8620396256953117e-07, "epoch": 0.9781790820165538, "percentage": 97.83, "elapsed_time": "2:15:17", "remaining_time": "0:02:59"}
+{"current_steps": 3260, "total_steps": 3322, "loss": 0.7348, "learning_rate": 2.122499551688084e-07, "epoch": 0.9811888638073739, "percentage": 98.13, "elapsed_time": "2:15:41", "remaining_time": "0:02:34"}
+{"current_steps": 3270, "total_steps": 3322, "loss": 0.7116, "learning_rate": 1.4931948942895624e-07, "epoch": 0.9841986455981941, "percentage": 98.43, "elapsed_time": "2:16:05", "remaining_time": "0:02:09"}
+{"current_steps": 3280, "total_steps": 3322, "loss": 0.7242, "learning_rate": 9.741951727152421e-08, "epoch": 0.9872084273890143, "percentage": 98.74, "elapsed_time": "2:16:28", "remaining_time": "0:01:44"}
+{"current_steps": 3290, "total_steps": 3322, "loss": 0.755, "learning_rate": 5.655577208069085e-08, "epoch": 0.9902182091798345, "percentage": 99.04, "elapsed_time": "2:16:51", "remaining_time": "0:01:19"}
+{"current_steps": 3300, "total_steps": 3322, "loss": 0.7697, "learning_rate": 2.6732768069825943e-08, "epoch": 0.9932279909706546, "percentage": 99.34, "elapsed_time": "2:17:16", "remaining_time": "0:00:54"}
+{"current_steps": 3310, "total_steps": 3322, "loss": 0.7166, "learning_rate": 7.953799782889349e-09, "epoch": 0.9962377727614747, "percentage": 99.64, "elapsed_time": "2:17:40", "remaining_time": "0:00:29"}
+{"current_steps": 3320, "total_steps": 3322, "loss": 0.7366, "learning_rate": 2.2094173039999277e-10, "epoch": 0.999247554552295, "percentage": 99.94, "elapsed_time": "2:18:03", "remaining_time": "0:00:04"}
+{"current_steps": 3322, "total_steps": 3322, "epoch": 0.999849510910459, "percentage": 100.0, "elapsed_time": "2:18:09", "remaining_time": "0:00:00"}

sharegpt/trainer_state.json

@@ -0,0 +1,2390 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.999849510910459,
+  "eval_steps": 1000,
+  "global_step": 3322,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0030097817908201654,
+      "grad_norm": 0.5666791796684265,
+      "learning_rate": 6.006006006006006e-06,
+      "loss": 1.0909,
+      "step": 10
+    },
+    {
+      "epoch": 0.006019563581640331,
+      "grad_norm": 0.5926806926727295,
+      "learning_rate": 1.2012012012012012e-05,
+      "loss": 1.0813,
+      "step": 20
+    },
+    {
+      "epoch": 0.009029345372460496,
+      "grad_norm": 0.5680918097496033,
+      "learning_rate": 1.801801801801802e-05,
+      "loss": 1.051,
+      "step": 30
+    },
+    {
+      "epoch": 0.012039127163280662,
+      "grad_norm": 0.4019787907600403,
+      "learning_rate": 2.4024024024024024e-05,
+      "loss": 0.9406,
+      "step": 40
+    },
+    {
+      "epoch": 0.015048908954100828,
+      "grad_norm": 0.289296418428421,
+      "learning_rate": 3.0030030030030033e-05,
+      "loss": 0.9232,
+      "step": 50
+    },
+    {
+      "epoch": 0.01805869074492099,
+      "grad_norm": 0.23350511491298676,
+      "learning_rate": 3.603603603603604e-05,
+      "loss": 0.8762,
+      "step": 60
+    },
+    {
+      "epoch": 0.021068472535741158,
+      "grad_norm": 0.17287714779376984,
+      "learning_rate": 4.204204204204204e-05,
+      "loss": 0.8496,
+      "step": 70
+    },
+    {
+      "epoch": 0.024078254326561323,
+      "grad_norm": 0.16750749945640564,
+      "learning_rate": 4.804804804804805e-05,
+      "loss": 0.8424,
+      "step": 80
+    },
+    {
+      "epoch": 0.02708803611738149,
+      "grad_norm": 0.15370824933052063,
+      "learning_rate": 5.405405405405406e-05,
+      "loss": 0.8601,
+      "step": 90
+    },
+    {
+      "epoch": 0.030097817908201655,
+      "grad_norm": 0.16766728460788727,
+      "learning_rate": 6.0060060060060066e-05,
+      "loss": 0.8179,
+      "step": 100
+    },
+    {
+      "epoch": 0.03310759969902182,
+      "grad_norm": 0.2106652557849884,
+      "learning_rate": 6.606606606606607e-05,
+      "loss": 0.8094,
+      "step": 110
+    },
+    {
+      "epoch": 0.03611738148984198,
+      "grad_norm": 0.22345463931560516,
+      "learning_rate": 7.207207207207208e-05,
+      "loss": 0.8096,
+      "step": 120
+    },
+    {
+      "epoch": 0.03912716328066215,
+      "grad_norm": 0.23186658322811127,
+      "learning_rate": 7.807807807807808e-05,
+      "loss": 0.8195,
+      "step": 130
+    },
+    {
+      "epoch": 0.042136945071482315,
+      "grad_norm": 0.3011253774166107,
+      "learning_rate": 8.408408408408409e-05,
+      "loss": 0.8184,
+      "step": 140
+    },
+    {
+      "epoch": 0.045146726862302484,
+      "grad_norm": 0.21401682496070862,
+      "learning_rate": 9.009009009009009e-05,
+      "loss": 0.8,
+      "step": 150
+    },
+    {
+      "epoch": 0.04815650865312265,
+      "grad_norm": 0.2243528664112091,
+      "learning_rate": 9.60960960960961e-05,
+      "loss": 0.7739,
+      "step": 160
+    },
+    {
+      "epoch": 0.051166290443942816,
+      "grad_norm": 0.46177345514297485,
+      "learning_rate": 0.00010210210210210212,
+      "loss": 0.8172,
+      "step": 170
+    },
+    {
+      "epoch": 0.05417607223476298,
+      "grad_norm": 0.2084399163722992,
+      "learning_rate": 0.00010810810810810812,
+      "loss": 0.8181,
+      "step": 180
+    },
+    {
+      "epoch": 0.05718585402558315,
+      "grad_norm": 0.24123747646808624,
+      "learning_rate": 0.00011411411411411413,
+      "loss": 0.8032,
+      "step": 190
+    },
+    {
+      "epoch": 0.06019563581640331,
+      "grad_norm": 0.23546789586544037,
+      "learning_rate": 0.00012012012012012013,
+      "loss": 0.7775,
+      "step": 200
+    },
+    {
+      "epoch": 0.06320541760722348,
+      "grad_norm": 0.23672956228256226,
+      "learning_rate": 0.00012612612612612612,
+      "loss": 0.821,
+      "step": 210
+    },
+    {
+      "epoch": 0.06621519939804364,
+      "grad_norm": 0.22630153596401215,
+      "learning_rate": 0.00013213213213213214,
+      "loss": 0.7892,
+      "step": 220
+    },
+    {
+      "epoch": 0.0692249811888638,
+      "grad_norm": 0.2221691757440567,
+      "learning_rate": 0.00013813813813813813,
+      "loss": 0.7932,
+      "step": 230
+    },
+    {
+      "epoch": 0.07223476297968397,
+      "grad_norm": 0.2853372097015381,
+      "learning_rate": 0.00014414414414414415,
+      "loss": 0.7993,
+      "step": 240
+    },
+    {
+      "epoch": 0.07524454477050414,
+      "grad_norm": 0.24529512226581573,
+      "learning_rate": 0.00015015015015015014,
+      "loss": 0.8281,
+      "step": 250
+    },
+    {
+      "epoch": 0.0782543265613243,
+      "grad_norm": 0.20747515559196472,
+      "learning_rate": 0.00015615615615615616,
+      "loss": 0.7695,
+      "step": 260
+    },
+    {
+      "epoch": 0.08126410835214447,
+      "grad_norm": 0.21736431121826172,
+      "learning_rate": 0.00016216216216216218,
+      "loss": 0.7863,
+      "step": 270
+    },
+    {
+      "epoch": 0.08427389014296463,
+      "grad_norm": 0.25259608030319214,
+      "learning_rate": 0.00016816816816816817,
+      "loss": 0.7616,
+      "step": 280
+    },
+    {
+      "epoch": 0.0872836719337848,
+      "grad_norm": 0.19296181201934814,
+      "learning_rate": 0.0001741741741741742,
+      "loss": 0.7707,
+      "step": 290
+    },
+    {
+      "epoch": 0.09029345372460497,
+      "grad_norm": 0.26890674233436584,
+      "learning_rate": 0.00018018018018018018,
+      "loss": 0.7744,
+      "step": 300
+    },
+    {
+      "epoch": 0.09330323551542513,
+      "grad_norm": 0.2053564041852951,
+      "learning_rate": 0.0001861861861861862,
+      "loss": 0.743,
+      "step": 310
+    },
+    {
+      "epoch": 0.0963130173062453,
+      "grad_norm": 0.20555047690868378,
+      "learning_rate": 0.0001921921921921922,
+      "loss": 0.8089,
+      "step": 320
+    },
+    {
+      "epoch": 0.09932279909706546,
+      "grad_norm": 0.2425510585308075,
+      "learning_rate": 0.0001981981981981982,
+      "loss": 0.8012,
+      "step": 330
+    },
+    {
+      "epoch": 0.10233258088788563,
+      "grad_norm": 0.2260735034942627,
+      "learning_rate": 0.00019999729347501484,
+      "loss": 0.7843,
+      "step": 340
+    },
+    {
+      "epoch": 0.1053423626787058,
+      "grad_norm": 0.21512670814990997,
+      "learning_rate": 0.0001999840373787939,
+      "loss": 0.7992,
+      "step": 350
+    },
+    {
+      "epoch": 0.10835214446952596,
+      "grad_norm": 0.22542227804660797,
+      "learning_rate": 0.0001999597360570722,
+      "loss": 0.772,
+      "step": 360
+    },
+    {
+      "epoch": 0.11136192626034612,
+      "grad_norm": 0.15299277007579803,
+      "learning_rate": 0.0001999243921944139,
+      "loss": 0.7511,
+      "step": 370
+    },
+    {
+      "epoch": 0.1143717080511663,
+      "grad_norm": 0.27176716923713684,
+      "learning_rate": 0.00019987800969525164,
+      "loss": 0.7735,
+      "step": 380
+    },
+    {
+      "epoch": 0.11738148984198646,
+      "grad_norm": 0.28357240557670593,
+      "learning_rate": 0.00019982059368345496,
+      "loss": 0.7729,
+      "step": 390
+    },
+    {
+      "epoch": 0.12039127163280662,
+      "grad_norm": 0.18754735589027405,
+      "learning_rate": 0.00019975215050176433,
+      "loss": 0.795,
+      "step": 400
+    },
+    {
+      "epoch": 0.12340105342362678,
+      "grad_norm": 0.15982478857040405,
+      "learning_rate": 0.00019967268771109035,
+      "loss": 0.7671,
+      "step": 410
+    },
+    {
+      "epoch": 0.12641083521444696,
+      "grad_norm": 0.18540535867214203,
+      "learning_rate": 0.00019958221408967875,
+      "loss": 0.7491,
+      "step": 420
+    },
+    {
+      "epoch": 0.1294206170052671,
+      "grad_norm": 0.4123195707798004,
+      "learning_rate": 0.00019948073963214043,
+      "loss": 0.7766,
+      "step": 430
+    },
+    {
+      "epoch": 0.13243039879608728,
+      "grad_norm": 0.1608065813779831,
+      "learning_rate": 0.00019936827554834738,
+      "loss": 0.7904,
+      "step": 440
+    },
+    {
+      "epoch": 0.13544018058690746,
+      "grad_norm": 0.21908819675445557,
+      "learning_rate": 0.00019924483426219452,
+      "loss": 0.7816,
+      "step": 450
+    },
+    {
+      "epoch": 0.1384499623777276,
+      "grad_norm": 0.17256338894367218,
+      "learning_rate": 0.00019911042941022695,
+      "loss": 0.7982,
+      "step": 460
+    },
+    {
+      "epoch": 0.14145974416854779,
+      "grad_norm": 0.15286709368228912,
+      "learning_rate": 0.00019896507584013376,
+      "loss": 0.793,
+      "step": 470
+    },
+    {
+      "epoch": 0.14446952595936793,
+      "grad_norm": 0.1949639767408371,
+      "learning_rate": 0.00019880878960910772,
+      "loss": 0.745,
+      "step": 480
+    },
+    {
+      "epoch": 0.1474793077501881,
+      "grad_norm": 0.22766350209712982,
+      "learning_rate": 0.00019864158798207137,
+      "loss": 0.765,
+      "step": 490
+    },
+    {
+      "epoch": 0.1504890895410083,
+      "grad_norm": 0.17246346175670624,
+      "learning_rate": 0.0001984634894297699,
+      "loss": 0.7494,
+      "step": 500
+    },
+    {
+      "epoch": 0.15349887133182843,
+      "grad_norm": 0.18809442222118378,
+      "learning_rate": 0.00019827451362673052,
+      "loss": 0.7906,
+      "step": 510
+    },
+    {
+      "epoch": 0.1565086531226486,
+      "grad_norm": 0.287610799074173,
+      "learning_rate": 0.00019807468144908928,
+      "loss": 0.7991,
+      "step": 520
+    },
+    {
+      "epoch": 0.1595184349134688,
+      "grad_norm": 0.2345225214958191,
+      "learning_rate": 0.00019786401497228466,
+      "loss": 0.7593,
+      "step": 530
+    },
+    {
+      "epoch": 0.16252821670428894,
+      "grad_norm": 0.24265146255493164,
+      "learning_rate": 0.00019764253746861886,
+      "loss": 0.7966,
+      "step": 540
+    },
+    {
+      "epoch": 0.1655379984951091,
+      "grad_norm": 0.1692132204771042,
+      "learning_rate": 0.00019741027340468715,
+      "loss": 0.7525,
+      "step": 550
+    },
+    {
+      "epoch": 0.16854778028592926,
+      "grad_norm": 0.18456414341926575,
+      "learning_rate": 0.00019716724843867487,
+      "loss": 0.7706,
+      "step": 560
+    },
+    {
+      "epoch": 0.17155756207674944,
+      "grad_norm": 0.2189481407403946,
+      "learning_rate": 0.000196913489417523,
+      "loss": 0.7683,
+      "step": 570
+    },
+    {
+      "epoch": 0.1745673438675696,
+      "grad_norm": 0.24484622478485107,
+      "learning_rate": 0.00019664902437396245,
+      "loss": 0.821,
+      "step": 580
+    },
+    {
+      "epoch": 0.17757712565838976,
+      "grad_norm": 0.1760026067495346,
+      "learning_rate": 0.00019637388252341715,
+      "loss": 0.7744,
+      "step": 590
+    },
+    {
+      "epoch": 0.18058690744920994,
+      "grad_norm": 0.18553385138511658,
+      "learning_rate": 0.00019608809426077678,
+      "loss": 0.7607,
+      "step": 600
+    },
+    {
+      "epoch": 0.1835966892400301,
+      "grad_norm": 0.23498745262622833,
+      "learning_rate": 0.000195791691157039,
+      "loss": 0.7707,
+      "step": 610
+    },
+    {
+      "epoch": 0.18660647103085026,
+      "grad_norm": 0.20103472471237183,
+      "learning_rate": 0.00019548470595582166,
+      "loss": 0.7487,
+      "step": 620
+    },
+    {
+      "epoch": 0.18961625282167044,
+      "grad_norm": 0.18829959630966187,
+      "learning_rate": 0.00019516717256974592,
+      "loss": 0.7653,
+      "step": 630
+    },
+    {
+      "epoch": 0.1926260346124906,
+      "grad_norm": 0.24024085700511932,
+      "learning_rate": 0.00019483912607668965,
+      "loss": 0.7918,
+      "step": 640
+    },
+    {
+      "epoch": 0.19563581640331076,
+      "grad_norm": 0.18580889701843262,
+      "learning_rate": 0.00019450060271591243,
+      "loss": 0.8022,
+      "step": 650
+    },
+    {
+      "epoch": 0.1986455981941309,
+      "grad_norm": 0.2515565752983093,
+      "learning_rate": 0.0001941516398840524,
+      "loss": 0.7548,
+      "step": 660
+    },
+    {
+      "epoch": 0.2016553799849511,
+      "grad_norm": 0.18456920981407166,
+      "learning_rate": 0.00019379227613099473,
+      "loss": 0.7903,
+      "step": 670
+    },
+    {
+      "epoch": 0.20466516177577126,
+      "grad_norm": 0.23653873801231384,
+      "learning_rate": 0.00019342255115561337,
+      "loss": 0.7917,
+      "step": 680
+    },
+    {
+      "epoch": 0.2076749435665914,
+      "grad_norm": 0.26416492462158203,
+      "learning_rate": 0.00019304250580138524,
+      "loss": 0.7879,
+      "step": 690
+    },
+    {
+      "epoch": 0.2106847253574116,
+      "grad_norm": 0.22712339460849762,
+      "learning_rate": 0.0001926521820518784,
+      "loss": 0.8014,
+      "step": 700
+    },
+    {
+      "epoch": 0.21369450714823177,
+      "grad_norm": 0.17389413714408875,
+      "learning_rate": 0.00019225162302611412,
+      "loss": 0.7403,
+      "step": 710
+    },
+    {
+      "epoch": 0.21670428893905191,
+      "grad_norm": 0.23703350126743317,
+      "learning_rate": 0.00019184087297380344,
+      "loss": 0.7795,
+      "step": 720
+    },
+    {
+      "epoch": 0.2197140707298721,
+      "grad_norm": 0.20507369935512543,
+      "learning_rate": 0.000191419977270459,
+      "loss": 0.8047,
+      "step": 730
+    },
+    {
+      "epoch": 0.22272385252069224,
+      "grad_norm": 0.1989058554172516,
+      "learning_rate": 0.0001909889824123824,
+      "loss": 0.7786,
+      "step": 740
+    },
+    {
+      "epoch": 0.22573363431151242,
+      "grad_norm": 0.1992356926202774,
+      "learning_rate": 0.00019054793601152773,
+      "loss": 0.7352,
+      "step": 750
+    },
+    {
+      "epoch": 0.2287434161023326,
+      "grad_norm": 0.1841665804386139,
+      "learning_rate": 0.0001900968867902419,
+      "loss": 0.778,
+      "step": 760
+    },
+    {
+      "epoch": 0.23175319789315274,
+      "grad_norm": 0.1729171723127365,
+      "learning_rate": 0.00018963588457588228,
+      "loss": 0.7776,
+      "step": 770
+    },
+    {
+      "epoch": 0.23476297968397292,
+      "grad_norm": 0.18624809384346008,
+      "learning_rate": 0.00018916498029531223,
+      "loss": 0.7841,
+      "step": 780
+    },
+    {
+      "epoch": 0.23777276147479307,
+      "grad_norm": 0.19757080078125,
+      "learning_rate": 0.00018868422596927535,
+      "loss": 0.7628,
+      "step": 790
+    },
+    {
+      "epoch": 0.24078254326561324,
+      "grad_norm": 0.1995578110218048,
+      "learning_rate": 0.00018819367470664862,
+      "loss": 0.7403,
+      "step": 800
+    },
+    {
+      "epoch": 0.24379232505643342,
+      "grad_norm": 0.2151431441307068,
+      "learning_rate": 0.00018769338069857548,
+      "loss": 0.7581,
+      "step": 810
+    },
+    {
+      "epoch": 0.24680210684725357,
+      "grad_norm": 0.2272290140390396,
+      "learning_rate": 0.00018718339921247945,
+      "loss": 0.7914,
+      "step": 820
+    },
+    {
+      "epoch": 0.24981188863807374,
+      "grad_norm": 0.1463918834924698,
+      "learning_rate": 0.0001866637865859586,
+      "loss": 0.7953,
+      "step": 830
+    },
+    {
+      "epoch": 0.2528216704288939,
+      "grad_norm": 0.2657776176929474,
+      "learning_rate": 0.00018613460022056215,
+      "loss": 0.7576,
+      "step": 840
+    },
+    {
+      "epoch": 0.2558314522197141,
+      "grad_norm": 0.2566029131412506,
+      "learning_rate": 0.000185595898575449,
+      "loss": 0.7508,
+      "step": 850
+    },
+    {
+      "epoch": 0.2588412340105342,
+      "grad_norm": 0.2356068342924118,
+      "learning_rate": 0.00018504774116093008,
+      "loss": 0.7332,
+      "step": 860
+    },
+    {
+      "epoch": 0.2618510158013544,
+      "grad_norm": 0.20745734870433807,
+      "learning_rate": 0.00018449018853189403,
+      "loss": 0.756,
+      "step": 870
+    },
+    {
+      "epoch": 0.26486079759217457,
+      "grad_norm": 0.1839013397693634,
+      "learning_rate": 0.0001839233022811179,
+      "loss": 0.7776,
+      "step": 880
+    },
+    {
+      "epoch": 0.26787057938299474,
+      "grad_norm": 0.238608717918396,
+      "learning_rate": 0.00018334714503246273,
+      "loss": 0.7771,
+      "step": 890
+    },
+    {
+      "epoch": 0.2708803611738149,
+      "grad_norm": 0.17599129676818848,
+      "learning_rate": 0.00018276178043395586,
+      "loss": 0.7544,
+      "step": 900
+    },
+    {
+      "epoch": 0.27389014296463504,
+      "grad_norm": 0.23767121136188507,
+      "learning_rate": 0.00018216727315075945,
+      "loss": 0.7946,
+      "step": 910
+    },
+    {
+      "epoch": 0.2768999247554552,
+      "grad_norm": 0.180665984749794,
+      "learning_rate": 0.00018156368885802695,
+      "loss": 0.8202,
+      "step": 920
+    },
+    {
+      "epoch": 0.2799097065462754,
+      "grad_norm": 0.20493340492248535,
+      "learning_rate": 0.00018095109423364817,
+      "loss": 0.7823,
+      "step": 930
+    },
+    {
+      "epoch": 0.28291948833709557,
+      "grad_norm": 0.29762348532676697,
+      "learning_rate": 0.0001803295569508832,
+      "loss": 0.7637,
+      "step": 940
+    },
+    {
+      "epoch": 0.28592927012791575,
+      "grad_norm": 0.2135084718465805,
+      "learning_rate": 0.0001796991456708866,
+      "loss": 0.768,
+      "step": 950
+    },
+    {
+      "epoch": 0.28893905191873587,
+      "grad_norm": 0.21105533838272095,
+      "learning_rate": 0.0001790599300351225,
+      "loss": 0.7492,
+      "step": 960
+    },
+    {
+      "epoch": 0.29194883370955604,
+      "grad_norm": 0.2163008451461792,
+      "learning_rate": 0.00017841198065767107,
+      "loss": 0.758,
+      "step": 970
+    },
+    {
+      "epoch": 0.2949586155003762,
+      "grad_norm": 0.16340641677379608,
+      "learning_rate": 0.00017775536911742806,
+      "loss": 0.7739,
+      "step": 980
+    },
+    {
+      "epoch": 0.2979683972911964,
+      "grad_norm": 0.20751433074474335,
+      "learning_rate": 0.00017709016795019742,
+      "loss": 0.7692,
+      "step": 990
+    },
+    {
+      "epoch": 0.3009781790820166,
+      "grad_norm": 0.1802573949098587,
+      "learning_rate": 0.00017641645064067816,
+      "loss": 0.7886,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3009781790820166,
+      "eval_loss": 0.8210044503211975,
+      "eval_runtime": 143.0609,
+      "eval_samples_per_second": 39.116,
+      "eval_steps_per_second": 4.893,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3039879608728367,
+      "grad_norm": 0.17218339443206787,
+      "learning_rate": 0.0001757342916143466,
+      "loss": 0.7595,
+      "step": 1010
+    },
+    {
+      "epoch": 0.30699774266365687,
+      "grad_norm": 0.3174282908439636,
+      "learning_rate": 0.00017504376622923465,
+      "loss": 0.7821,
+      "step": 1020
+    },
+    {
+      "epoch": 0.31000752445447705,
+      "grad_norm": 0.1862519532442093,
+      "learning_rate": 0.00017434495076760483,
+      "loss": 0.7982,
+      "step": 1030
+    },
+    {
+      "epoch": 0.3130173062452972,
+      "grad_norm": 0.19561271369457245,
+      "learning_rate": 0.00017363792242752353,
+      "loss": 0.7422,
+      "step": 1040
+    },
+    {
+      "epoch": 0.3160270880361174,
+      "grad_norm": 0.24750228226184845,
+      "learning_rate": 0.000172922759314333,
+      "loss": 0.7425,
+      "step": 1050
+    },
+    {
+      "epoch": 0.3190368698269376,
+      "grad_norm": 0.22299301624298096,
+      "learning_rate": 0.0001721995404320228,
+      "loss": 0.7392,
+      "step": 1060
+    },
+    {
+      "epoch": 0.3220466516177577,
+      "grad_norm": 0.38551756739616394,
+      "learning_rate": 0.0001714683456745026,
+      "loss": 0.7913,
+      "step": 1070
+    },
+    {
+      "epoch": 0.32505643340857787,
+      "grad_norm": 0.19068562984466553,
+      "learning_rate": 0.00017072925581677594,
+      "loss": 0.7368,
+      "step": 1080
+    },
+    {
+      "epoch": 0.32806621519939805,
+      "grad_norm": 0.14717283844947815,
+      "learning_rate": 0.0001699823525060174,
+      "loss": 0.7938,
+      "step": 1090
+    },
+    {
+      "epoch": 0.3310759969902182,
+      "grad_norm": 0.20433548092842102,
+      "learning_rate": 0.00016922771825255263,
+      "loss": 0.7672,
+      "step": 1100
+    },
+    {
+      "epoch": 0.3340857787810384,
+      "grad_norm": 0.24607650935649872,
+      "learning_rate": 0.0001684654364207438,
+      "loss": 0.7971,
+      "step": 1110
+    },
+    {
+      "epoch": 0.3370955605718585,
+      "grad_norm": 0.1704825758934021,
+      "learning_rate": 0.00016769559121978026,
+      "loss": 0.7283,
+      "step": 1120
+    },
+    {
+      "epoch": 0.3401053423626787,
+      "grad_norm": 0.18218447268009186,
+      "learning_rate": 0.0001669182676943757,
+      "loss": 0.7405,
+      "step": 1130
+    },
+    {
+      "epoch": 0.3431151241534989,
+      "grad_norm": 0.25802308320999146,
+      "learning_rate": 0.0001661335517153737,
+      "loss": 0.7821,
+      "step": 1140
+    },
+    {
+      "epoch": 0.34612490594431905,
+      "grad_norm": 0.2192961871623993,
+      "learning_rate": 0.00016534152997026125,
+      "loss": 0.7392,
+      "step": 1150
+    },
+    {
+      "epoch": 0.3491346877351392,
+      "grad_norm": 0.19438660144805908,
+      "learning_rate": 0.00016454228995359252,
+      "loss": 0.7928,
+      "step": 1160
+    },
+    {
+      "epoch": 0.35214446952595935,
+      "grad_norm": 0.16023948788642883,
+      "learning_rate": 0.00016373591995732338,
+      "loss": 0.7542,
+      "step": 1170
+    },
+    {
+      "epoch": 0.3551542513167795,
+      "grad_norm": 0.32932865619659424,
+      "learning_rate": 0.0001629225090610577,
+      "loss": 0.7766,
+      "step": 1180
+    },
+    {
+      "epoch": 0.3581640331075997,
+      "grad_norm": 0.282248854637146,
+      "learning_rate": 0.00016210214712220687,
+      "loss": 0.7528,
+      "step": 1190
+    },
+    {
+      "epoch": 0.3611738148984199,
+      "grad_norm": 0.2124871164560318,
+      "learning_rate": 0.00016127492476606308,
+      "loss": 0.7874,
+      "step": 1200
+    },
+    {
+      "epoch": 0.36418359668924005,
+      "grad_norm": 0.23827847838401794,
+      "learning_rate": 0.00016044093337578815,
+      "loss": 0.7599,
+      "step": 1210
+    },
+    {
+      "epoch": 0.3671933784800602,
+      "grad_norm": 0.22505390644073486,
+      "learning_rate": 0.00015960026508231824,
+      "loss": 0.7707,
+      "step": 1220
+    },
+    {
+      "epoch": 0.37020316027088035,
+      "grad_norm": 0.17258504033088684,
+      "learning_rate": 0.00015875301275418638,
+      "loss": 0.8102,
+      "step": 1230
+    },
+    {
+      "epoch": 0.3732129420617005,
+      "grad_norm": 0.24378369748592377,
+      "learning_rate": 0.00015789926998726315,
+      "loss": 0.7388,
+      "step": 1240
+    },
+    {
+      "epoch": 0.3762227238525207,
+      "grad_norm": 0.22882601618766785,
+      "learning_rate": 0.00015703913109441713,
+      "loss": 0.7707,
+      "step": 1250
+    },
+    {
+      "epoch": 0.3792325056433409,
+      "grad_norm": 0.20203706622123718,
+      "learning_rate": 0.0001561726910950962,
+      "loss": 0.7444,
+      "step": 1260
+    },
+    {
+      "epoch": 0.382242287434161,
+      "grad_norm": 0.25836071372032166,
+      "learning_rate": 0.00015530004570483093,
+      "loss": 0.7838,
+      "step": 1270
+    },
+    {
+      "epoch": 0.3852520692249812,
+      "grad_norm": 0.16395032405853271,
+      "learning_rate": 0.00015442129132466054,
+      "loss": 0.7281,
+      "step": 1280
+    },
+    {
+      "epoch": 0.38826185101580135,
+      "grad_norm": 0.19731546938419342,
+      "learning_rate": 0.00015353652503048384,
+      "loss": 0.7471,
+      "step": 1290
+    },
+    {
+      "epoch": 0.3912716328066215,
+      "grad_norm": 0.23747088015079498,
+      "learning_rate": 0.00015264584456233502,
+      "loss": 0.7469,
+      "step": 1300
+    },
+    {
+      "epoch": 0.3942814145974417,
+      "grad_norm": 0.222218319773674,
+      "learning_rate": 0.0001517493483135864,
+      "loss": 0.7945,
+      "step": 1310
+    },
+    {
+      "epoch": 0.3972911963882618,
+      "grad_norm": 0.21308743953704834,
+      "learning_rate": 0.00015084713532007905,
+      "loss": 0.7659,
+      "step": 1320
+    },
+    {
+      "epoch": 0.400300978179082,
+      "grad_norm": 0.25059592723846436,
+      "learning_rate": 0.00014993930524918208,
+      "loss": 0.7718,
+      "step": 1330
+    },
+    {
+      "epoch": 0.4033107599699022,
+      "grad_norm": 0.2029498815536499,
+      "learning_rate": 0.00014902595838878256,
+      "loss": 0.759,
+      "step": 1340
+    },
+    {
+      "epoch": 0.40632054176072235,
+      "grad_norm": 0.26108860969543457,
+      "learning_rate": 0.0001481071956362067,
+      "loss": 0.7568,
+      "step": 1350
+    },
+    {
+      "epoch": 0.40933032355154253,
+      "grad_norm": 0.16724595427513123,
+      "learning_rate": 0.0001471831184870737,
+      "loss": 0.7504,
+      "step": 1360
+    },
+    {
+      "epoch": 0.4123401053423627,
+      "grad_norm": 0.2351473867893219,
+      "learning_rate": 0.00014625382902408356,
+      "loss": 0.7365,
+      "step": 1370
+    },
+    {
+      "epoch": 0.4153498871331828,
+      "grad_norm": 0.20350322127342224,
+      "learning_rate": 0.00014531942990573998,
+      "loss": 0.7444,
+      "step": 1380
+    },
+    {
+      "epoch": 0.418359668924003,
+      "grad_norm": 0.2595962882041931,
+      "learning_rate": 0.00014438002435500979,
+      "loss": 0.7574,
+      "step": 1390
+    },
+    {
+      "epoch": 0.4213694507148232,
+      "grad_norm": 0.28216540813446045,
+      "learning_rate": 0.0001434357161479198,
+      "loss": 0.74,
+      "step": 1400
+    },
+    {
+      "epoch": 0.42437923250564336,
+      "grad_norm": 0.19514194130897522,
+      "learning_rate": 0.0001424866096020927,
+      "loss": 0.7761,
+      "step": 1410
+    },
+    {
+      "epoch": 0.42738901429646353,
+      "grad_norm": 0.2227347493171692,
+      "learning_rate": 0.00014153280956522322,
+      "loss": 0.7895,
+      "step": 1420
+    },
+    {
+      "epoch": 0.43039879608728365,
+      "grad_norm": 0.21557843685150146,
+      "learning_rate": 0.00014057442140349543,
+      "loss": 0.794,
+      "step": 1430
+    },
+    {
+      "epoch": 0.43340857787810383,
+      "grad_norm": 0.1980540156364441,
+      "learning_rate": 0.00013961155098994309,
+      "loss": 0.7471,
+      "step": 1440
+    },
+    {
+      "epoch": 0.436418359668924,
+      "grad_norm": 0.2989167869091034,
+      "learning_rate": 0.00013864430469275377,
+      "loss": 0.745,
+      "step": 1450
+    },
+    {
+      "epoch": 0.4394281414597442,
+      "grad_norm": 0.2816371023654938,
+      "learning_rate": 0.00013767278936351854,
+      "loss": 0.7392,
+      "step": 1460
+    },
+    {
+      "epoch": 0.44243792325056436,
+      "grad_norm": 0.20301824808120728,
+      "learning_rate": 0.00013669711232542776,
+      "loss": 0.7486,
+      "step": 1470
+    },
+    {
+      "epoch": 0.4454477050413845,
+      "grad_norm": 0.18164554238319397,
+      "learning_rate": 0.00013571738136141555,
+      "loss": 0.7571,
+      "step": 1480
+    },
+    {
+      "epoch": 0.44845748683220465,
+      "grad_norm": 0.20167161524295807,
+      "learning_rate": 0.0001347337047022526,
+      "loss": 0.76,
+      "step": 1490
+    },
+    {
+      "epoch": 0.45146726862302483,
+      "grad_norm": 0.18913504481315613,
+      "learning_rate": 0.00013374619101459012,
+      "loss": 0.7444,
+      "step": 1500
+    },
+    {
+      "epoch": 0.454477050413845,
+      "grad_norm": 0.2532965838909149,
+      "learning_rate": 0.00013275494938895556,
+      "loss": 0.7755,
+      "step": 1510
+    },
+    {
+      "epoch": 0.4574868322046652,
+      "grad_norm": 0.16350123286247253,
+      "learning_rate": 0.00013176008932770113,
+      "loss": 0.755,
+      "step": 1520
+    },
+    {
+      "epoch": 0.4604966139954853,
+      "grad_norm": 0.18730787932872772,
+      "learning_rate": 0.00013076172073290724,
+      "loss": 0.7369,
+      "step": 1530
+    },
+    {
+      "epoch": 0.4635063957863055,
+      "grad_norm": 0.21689856052398682,
+      "learning_rate": 0.00012975995389424166,
+      "loss": 0.7773,
+      "step": 1540
+    },
+    {
+      "epoch": 0.46651617757712566,
+      "grad_norm": 0.21407072246074677,
+      "learning_rate": 0.0001287548994767758,
+      "loss": 0.7287,
+      "step": 1550
+    },
+    {
+      "epoch": 0.46952595936794583,
+      "grad_norm": 0.2083187699317932,
+      "learning_rate": 0.00012774666850875942,
+      "loss": 0.7717,
+      "step": 1560
+    },
+    {
+      "epoch": 0.472535741158766,
+      "grad_norm": 0.2223493456840515,
+      "learning_rate": 0.00012673537236935556,
+      "loss": 0.7613,
+      "step": 1570
+    },
+    {
+      "epoch": 0.47554552294958613,
+      "grad_norm": 0.2172088325023651,
+      "learning_rate": 0.00012572112277633649,
+      "loss": 0.7602,
+      "step": 1580
+    },
+    {
+      "epoch": 0.4785553047404063,
+      "grad_norm": 0.22197513282299042,
+      "learning_rate": 0.0001247040317737419,
+      "loss": 0.7241,
+      "step": 1590
+    },
+    {
+      "epoch": 0.4815650865312265,
+      "grad_norm": 0.1837187558412552,
+      "learning_rate": 0.00012368421171950192,
+      "loss": 0.7313,
+      "step": 1600
+    },
+    {
+      "epoch": 0.48457486832204666,
+      "grad_norm": 0.23295271396636963,
+      "learning_rate": 0.00012266177527302472,
+      "loss": 0.7432,
+      "step": 1610
+    },
+    {
+      "epoch": 0.48758465011286684,
+      "grad_norm": 0.21737614274024963,
+      "learning_rate": 0.0001216368353827508,
+      "loss": 0.7599,
+      "step": 1620
+    },
+    {
+      "epoch": 0.49059443190368696,
+      "grad_norm": 0.1803632527589798,
+      "learning_rate": 0.00012060950527367603,
+      "loss": 0.7386,
+      "step": 1630
+    },
+    {
+      "epoch": 0.49360421369450713,
+      "grad_norm": 0.1795688271522522,
+      "learning_rate": 0.00011957989843484345,
+      "loss": 0.7548,
+      "step": 1640
+    },
+    {
+      "epoch": 0.4966139954853273,
+      "grad_norm": 0.2615033984184265,
+      "learning_rate": 0.00011854812860680613,
+      "loss": 0.7838,
+      "step": 1650
+    },
+    {
+      "epoch": 0.4996237772761475,
+      "grad_norm": 0.22325466573238373,
+      "learning_rate": 0.00011751430976906233,
+      "loss": 0.7492,
+      "step": 1660
+    },
+    {
+      "epoch": 0.5026335590669676,
+      "grad_norm": 0.20187042653560638,
+      "learning_rate": 0.00011647855612746423,
+      "loss": 0.7897,
+      "step": 1670
+    },
+    {
+      "epoch": 0.5056433408577878,
+      "grad_norm": 0.3646783232688904,
+      "learning_rate": 0.00011544098210160152,
+      "loss": 0.7847,
+      "step": 1680
+    },
+    {
+      "epoch": 0.508653122648608,
+      "grad_norm": 0.23362590372562408,
+      "learning_rate": 0.00011440170231216154,
+      "loss": 0.7624,
+      "step": 1690
+    },
+    {
+      "epoch": 0.5116629044394282,
+      "grad_norm": 0.2684471011161804,
+      "learning_rate": 0.00011336083156826722,
+      "loss": 0.7973,
+      "step": 1700
+    },
+    {
+      "epoch": 0.5146726862302483,
+      "grad_norm": 0.15820720791816711,
+      "learning_rate": 0.00011231848485479395,
+      "loss": 0.7322,
+      "step": 1710
+    },
+    {
+      "epoch": 0.5176824680210684,
+      "grad_norm": 0.3005324900150299,
+      "learning_rate": 0.00011127477731966735,
+      "loss": 0.7449,
+      "step": 1720
+    },
+    {
+      "epoch": 0.5206922498118887,
+      "grad_norm": 0.19756928086280823,
+      "learning_rate": 0.00011022982426114292,
+      "loss": 0.7988,
+      "step": 1730
+    },
+    {
+      "epoch": 0.5237020316027088,
+      "grad_norm": 0.20102205872535706,
+      "learning_rate": 0.00010918374111506893,
+      "loss": 0.7273,
+      "step": 1740
+    },
+    {
+      "epoch": 0.526711813393529,
+      "grad_norm": 0.2193961888551712,
+      "learning_rate": 0.00010813664344213427,
+      "loss": 0.7367,
+      "step": 1750
+    },
+    {
+      "epoch": 0.5297215951843491,
+      "grad_norm": 0.2881150245666504,
+      "learning_rate": 0.00010708864691510254,
+      "loss": 0.7702,
+      "step": 1760
+    },
+    {
+      "epoch": 0.5327313769751693,
+      "grad_norm": 0.24800467491149902,
+      "learning_rate": 0.00010603986730603368,
+      "loss": 0.7853,
+      "step": 1770
+    },
+    {
+      "epoch": 0.5357411587659895,
+      "grad_norm": 0.18105538189411163,
+      "learning_rate": 0.00010499042047349455,
+      "loss": 0.7576,
+      "step": 1780
+    },
+    {
+      "epoch": 0.5387509405568096,
+      "grad_norm": 0.16151605546474457,
+      "learning_rate": 0.00010394042234976016,
+      "loss": 0.7363,
+      "step": 1790
+    },
+    {
+      "epoch": 0.5417607223476298,
+      "grad_norm": 0.2590145170688629,
+      "learning_rate": 0.00010288998892800657,
+      "loss": 0.7501,
+      "step": 1800
+    },
+    {
+      "epoch": 0.54477050413845,
+      "grad_norm": 0.22977079451084137,
+      "learning_rate": 0.0001018392362494972,
+      "loss": 0.7768,
+      "step": 1810
+    },
+    {
+      "epoch": 0.5477802859292701,
+      "grad_norm": 0.23535679280757904,
+      "learning_rate": 0.00010078828039076367,
+      "loss": 0.7803,
+      "step": 1820
+    },
+    {
+      "epoch": 0.5507900677200903,
+      "grad_norm": 0.43004941940307617,
+      "learning_rate": 9.973723745078296e-05,
+      "loss": 0.7686,
+      "step": 1830
+    },
+    {
+      "epoch": 0.5537998495109104,
+      "grad_norm": 0.25901973247528076,
+      "learning_rate": 9.868622353815188e-05,
+      "loss": 0.7623,
+      "step": 1840
+    },
+    {
+      "epoch": 0.5568096313017307,
+      "grad_norm": 0.2264987975358963,
+      "learning_rate": 9.763535475826054e-05,
+      "loss": 0.7439,
+      "step": 1850
+    },
+    {
+      "epoch": 0.5598194130925508,
+      "grad_norm": 0.24374960362911224,
+      "learning_rate": 9.658474720046637e-05,
+      "loss": 0.7825,
+      "step": 1860
+    },
+    {
+      "epoch": 0.5628291948833709,
+      "grad_norm": 0.20084655284881592,
+      "learning_rate": 9.553451692526954e-05,
+      "loss": 0.7802,
+      "step": 1870
+    },
+    {
+      "epoch": 0.5658389766741911,
+      "grad_norm": 0.1831783950328827,
+      "learning_rate": 9.448477995149182e-05,
+      "loss": 0.7328,
+      "step": 1880
+    },
+    {
+      "epoch": 0.5688487584650113,
+      "grad_norm": 0.18079884350299835,
+      "learning_rate": 9.343565224346013e-05,
+      "loss": 0.7464,
+      "step": 1890
+    },
+    {
+      "epoch": 0.5718585402558315,
+      "grad_norm": 0.248003751039505,
+      "learning_rate": 9.238724969819579e-05,
+      "loss": 0.7387,
+      "step": 1900
+    },
+    {
+      "epoch": 0.5748683220466516,
+      "grad_norm": 0.18932189047336578,
+      "learning_rate": 9.13396881326115e-05,
+      "loss": 0.7136,
+      "step": 1910
+    },
+    {
+      "epoch": 0.5778781038374717,
+      "grad_norm": 0.20788271725177765,
+      "learning_rate": 9.029308327071702e-05,
+      "loss": 0.7702,
+      "step": 1920
+    },
+    {
+      "epoch": 0.580887885628292,
+      "grad_norm": 0.19441328942775726,
+      "learning_rate": 8.924755073083517e-05,
+      "loss": 0.7901,
+      "step": 1930
+    },
+    {
+      "epoch": 0.5838976674191121,
+      "grad_norm": 0.21718983352184296,
+      "learning_rate": 8.820320601282949e-05,
+      "loss": 0.7771,
+      "step": 1940
+    },
+    {
+      "epoch": 0.5869074492099323,
+      "grad_norm": 0.19140474498271942,
+      "learning_rate": 8.71601644853449e-05,
+      "loss": 0.7601,
+      "step": 1950
+    },
+    {
+      "epoch": 0.5899172310007524,
+      "grad_norm": 0.26861098408699036,
+      "learning_rate": 8.61185413730631e-05,
+      "loss": 0.7575,
+      "step": 1960
+    },
+    {
+      "epoch": 0.5929270127915726,
+      "grad_norm": 0.19967325031757355,
+      "learning_rate": 8.507845174397357e-05,
+      "loss": 0.8136,
+      "step": 1970
+    },
+    {
+      "epoch": 0.5959367945823928,
+      "grad_norm": 0.19900915026664734,
+      "learning_rate": 8.404001049666211e-05,
+      "loss": 0.7205,
+      "step": 1980
+    },
+    {
+      "epoch": 0.5989465763732129,
+      "grad_norm": 0.25308361649513245,
+      "learning_rate": 8.300333234761787e-05,
+      "loss": 0.768,
+      "step": 1990
+    },
+    {
+      "epoch": 0.6019563581640331,
+      "grad_norm": 0.16838368773460388,
+      "learning_rate": 8.196853181856081e-05,
+      "loss": 0.7502,
+      "step": 2000
+    },
+    {
+      "epoch": 0.6019563581640331,
+      "eval_loss": 0.8055371046066284,
+      "eval_runtime": 143.0386,
+      "eval_samples_per_second": 39.122,
+      "eval_steps_per_second": 4.894,
+      "step": 2000
+    },
+    {
+      "epoch": 0.6049661399548533,
+      "grad_norm": 0.22218653559684753,
+      "learning_rate": 8.093572322379045e-05,
+      "loss": 0.7597,
+      "step": 2010
+    },
+    {
+      "epoch": 0.6079759217456734,
+      "grad_norm": 0.19360238313674927,
+      "learning_rate": 7.990502065755748e-05,
+      "loss": 0.7592,
+      "step": 2020
+    },
+    {
+      "epoch": 0.6109857035364936,
+      "grad_norm": 0.25584983825683594,
+      "learning_rate": 7.887653798145987e-05,
+      "loss": 0.7432,
+      "step": 2030
+    },
+    {
+      "epoch": 0.6139954853273137,
+      "grad_norm": 0.23388119041919708,
+      "learning_rate": 7.785038881186462e-05,
+      "loss": 0.7638,
+      "step": 2040
+    },
+    {
+      "epoch": 0.617005267118134,
+      "grad_norm": 0.24461683630943298,
+      "learning_rate": 7.682668650735645e-05,
+      "loss": 0.7451,
+      "step": 2050
+    },
+    {
+      "epoch": 0.6200150489089541,
+      "grad_norm": 0.20549526810646057,
+      "learning_rate": 7.580554415621522e-05,
+      "loss": 0.7432,
+      "step": 2060
+    },
+    {
+      "epoch": 0.6230248306997742,
+      "grad_norm": 0.22572945058345795,
+      "learning_rate": 7.478707456392302e-05,
+      "loss": 0.7924,
+      "step": 2070
+    },
+    {
+      "epoch": 0.6260346124905944,
+      "grad_norm": 0.33990249037742615,
+      "learning_rate": 7.377139024070254e-05,
+      "loss": 0.7471,
+      "step": 2080
+    },
+    {
+      "epoch": 0.6290443942814146,
+      "grad_norm": 0.2138250321149826,
+      "learning_rate": 7.275860338908815e-05,
+      "loss": 0.7535,
+      "step": 2090
+    },
+    {
+      "epoch": 0.6320541760722348,
+      "grad_norm": 0.17337530851364136,
+      "learning_rate": 7.174882589153076e-05,
+      "loss": 0.7523,
+      "step": 2100
+    },
+    {
+      "epoch": 0.6350639578630549,
+      "grad_norm": 0.2286282777786255,
+      "learning_rate": 7.07421692980384e-05,
+      "loss": 0.7507,
+      "step": 2110
+    },
+    {
+      "epoch": 0.6380737396538751,
+      "grad_norm": 0.19162622094154358,
+      "learning_rate": 6.973874481385312e-05,
+      "loss": 0.7671,
+      "step": 2120
+    },
+    {
+      "epoch": 0.6410835214446953,
+      "grad_norm": 0.22725722193717957,
+      "learning_rate": 6.873866328716614e-05,
+      "loss": 0.7595,
+      "step": 2130
+    },
+    {
+      "epoch": 0.6440933032355154,
+      "grad_norm": 0.28031495213508606,
+      "learning_rate": 6.774203519687265e-05,
+      "loss": 0.7384,
+      "step": 2140
+    },
+    {
+      "epoch": 0.6471030850263356,
+      "grad_norm": 0.31639155745506287,
+      "learning_rate": 6.674897064036706e-05,
+      "loss": 0.7629,
+      "step": 2150
+    },
+    {
+      "epoch": 0.6501128668171557,
+      "grad_norm": 0.2253294438123703,
+      "learning_rate": 6.575957932138057e-05,
+      "loss": 0.7793,
+      "step": 2160
+    },
+    {
+      "epoch": 0.653122648607976,
+      "grad_norm": 0.2402772158384323,
+      "learning_rate": 6.47739705378623e-05,
+      "loss": 0.7586,
+      "step": 2170
+    },
+    {
+      "epoch": 0.6561324303987961,
+      "grad_norm": 0.1558917760848999,
+      "learning_rate": 6.379225316990505e-05,
+      "loss": 0.7564,
+      "step": 2180
+    },
+    {
+      "epoch": 0.6591422121896162,
+      "grad_norm": 0.17613032460212708,
+      "learning_rate": 6.281453566771735e-05,
+      "loss": 0.7354,
+      "step": 2190
+    },
+    {
+      "epoch": 0.6621519939804364,
+      "grad_norm": 0.28283971548080444,
+      "learning_rate": 6.184092603964308e-05,
+      "loss": 0.8203,
+      "step": 2200
+    },
+    {
+      "epoch": 0.6651617757712566,
+      "grad_norm": 0.2295006364583969,
+      "learning_rate": 6.087153184022969e-05,
+      "loss": 0.7466,
+      "step": 2210
+    },
+    {
+      "epoch": 0.6681715575620768,
+      "grad_norm": 0.24166414141654968,
+      "learning_rate": 5.990646015834668e-05,
+      "loss": 0.76,
+      "step": 2220
+    },
+    {
+      "epoch": 0.6711813393528969,
+      "grad_norm": 0.5943595170974731,
+      "learning_rate": 5.894581760535549e-05,
+      "loss": 0.7662,
+      "step": 2230
+    },
+    {
+      "epoch": 0.674191121143717,
+      "grad_norm": 0.23785801231861115,
+      "learning_rate": 5.798971030333227e-05,
+      "loss": 0.7742,
+      "step": 2240
+    },
+    {
+      "epoch": 0.6772009029345373,
+      "grad_norm": 0.20206815004348755,
+      "learning_rate": 5.703824387334442e-05,
+      "loss": 0.7837,
+      "step": 2250
+    },
+    {
+      "epoch": 0.6802106847253574,
+      "grad_norm": 0.21082885563373566,
+      "learning_rate": 5.609152342378278e-05,
+      "loss": 0.758,
+      "step": 2260
+    },
+    {
+      "epoch": 0.6832204665161776,
+      "grad_norm": 0.1953180432319641,
+      "learning_rate": 5.514965353875019e-05,
+      "loss": 0.726,
+      "step": 2270
+    },
+    {
+      "epoch": 0.6862302483069977,
+      "grad_norm": 0.22035697102546692,
+      "learning_rate": 5.4212738266508245e-05,
+      "loss": 0.7199,
+      "step": 2280
+    },
+    {
+      "epoch": 0.6892400300978179,
+      "grad_norm": 0.17208510637283325,
+      "learning_rate": 5.3280881107982946e-05,
+      "loss": 0.7379,
+      "step": 2290
+    },
+    {
+      "epoch": 0.6922498118886381,
+      "grad_norm": 0.18434813618659973,
+      "learning_rate": 5.235418500533109e-05,
+      "loss": 0.7574,
+      "step": 2300
+    },
+    {
+      "epoch": 0.6952595936794582,
+      "grad_norm": 0.35970667004585266,
+      "learning_rate": 5.143275233056817e-05,
+      "loss": 0.7737,
+      "step": 2310
+    },
+    {
+      "epoch": 0.6982693754702785,
+      "grad_norm": 0.19193576276302338,
+      "learning_rate": 5.051668487425938e-05,
+      "loss": 0.7268,
+      "step": 2320
+    },
+    {
+      "epoch": 0.7012791572610986,
+      "grad_norm": 0.21549300849437714,
+      "learning_rate": 4.960608383427481e-05,
+      "loss": 0.7542,
+      "step": 2330
+    },
+    {
+      "epoch": 0.7042889390519187,
+      "grad_norm": 0.1781996786594391,
+      "learning_rate": 4.8701049804610265e-05,
+      "loss": 0.7493,
+      "step": 2340
+    },
+    {
+      "epoch": 0.7072987208427389,
+      "grad_norm": 0.17246761918067932,
+      "learning_rate": 4.780168276427441e-05,
+      "loss": 0.7468,
+      "step": 2350
+    },
+    {
+      "epoch": 0.710308502633559,
+      "grad_norm": 0.22511474788188934,
+      "learning_rate": 4.6908082066244275e-05,
+      "loss": 0.7667,
+      "step": 2360
+    },
+    {
+      "epoch": 0.7133182844243793,
+      "grad_norm": 0.22506214678287506,
+      "learning_rate": 4.602034642648968e-05,
+      "loss": 0.788,
+      "step": 2370
+    },
+    {
+      "epoch": 0.7163280662151994,
+      "grad_norm": 0.23567315936088562,
+      "learning_rate": 4.513857391306812e-05,
+      "loss": 0.772,
+      "step": 2380
+    },
+    {
+      "epoch": 0.7193378480060195,
+      "grad_norm": 0.21594233810901642,
+      "learning_rate": 4.4262861935291144e-05,
+      "loss": 0.7257,
+      "step": 2390
+    },
+    {
+      "epoch": 0.7223476297968398,
+      "grad_norm": 0.2095227688550949,
+      "learning_rate": 4.339330723296373e-05,
+      "loss": 0.7456,
+      "step": 2400
+    },
+    {
+      "epoch": 0.7253574115876599,
+      "grad_norm": 0.2402193695306778,
+      "learning_rate": 4.25300058656972e-05,
+      "loss": 0.7722,
+      "step": 2410
+    },
+    {
+      "epoch": 0.7283671933784801,
+      "grad_norm": 0.2426275908946991,
+      "learning_rate": 4.1673053202297676e-05,
+      "loss": 0.7871,
+      "step": 2420
+    },
+    {
+      "epoch": 0.7313769751693002,
+      "grad_norm": 0.1813143640756607,
+      "learning_rate": 4.0822543910230674e-05,
+      "loss": 0.7394,
+      "step": 2430
+    },
+    {
+      "epoch": 0.7343867569601203,
+      "grad_norm": 0.29781273007392883,
+      "learning_rate": 3.997857194516319e-05,
+      "loss": 0.7318,
+      "step": 2440
+    },
+    {
+      "epoch": 0.7373965387509406,
+      "grad_norm": 0.2380327582359314,
+      "learning_rate": 3.914123054058446e-05,
+      "loss": 0.7711,
+      "step": 2450
+    },
+    {
+      "epoch": 0.7404063205417607,
+      "grad_norm": 0.2532583773136139,
+      "learning_rate": 3.831061219750636e-05,
+      "loss": 0.7341,
+      "step": 2460
+    },
+    {
+      "epoch": 0.7434161023325809,
+      "grad_norm": 0.18625672161579132,
+      "learning_rate": 3.7486808674245047e-05,
+      "loss": 0.7349,
+      "step": 2470
+    },
+    {
+      "epoch": 0.746425884123401,
+      "grad_norm": 0.2749151289463043,
+      "learning_rate": 3.666991097628416e-05,
+      "loss": 0.7551,
+      "step": 2480
+    },
+    {
+      "epoch": 0.7494356659142212,
+      "grad_norm": 0.28592735528945923,
+      "learning_rate": 3.586000934622166e-05,
+      "loss": 0.7485,
+      "step": 2490
+    },
+    {
+      "epoch": 0.7524454477050414,
+      "grad_norm": 0.22789210081100464,
+      "learning_rate": 3.5057193253800624e-05,
+      "loss": 0.7308,
+      "step": 2500
+    },
+    {
+      "epoch": 0.7554552294958615,
+      "grad_norm": 0.2588096559047699,
+      "learning_rate": 3.426155138602558e-05,
+      "loss": 0.7717,
+      "step": 2510
+    },
+    {
+      "epoch": 0.7584650112866818,
+      "grad_norm": 0.2197035402059555,
+      "learning_rate": 3.347317163736524e-05,
+      "loss": 0.753,
+      "step": 2520
+    },
+    {
+      "epoch": 0.7614747930775019,
+      "grad_norm": 0.19545480608940125,
+      "learning_rate": 3.269214110004293e-05,
+      "loss": 0.7552,
+      "step": 2530
+    },
+    {
+      "epoch": 0.764484574868322,
+      "grad_norm": 0.1901889145374298,
+      "learning_rate": 3.191854605441527e-05,
+      "loss": 0.7146,
+      "step": 2540
+    },
+    {
+      "epoch": 0.7674943566591422,
+      "grad_norm": 0.19916728138923645,
+      "learning_rate": 3.115247195944102e-05,
+      "loss": 0.7733,
+      "step": 2550
+    },
+    {
+      "epoch": 0.7705041384499624,
+      "grad_norm": 0.2248535454273224,
+      "learning_rate": 3.039400344324035e-05,
+      "loss": 0.7948,
+      "step": 2560
+    },
+    {
+      "epoch": 0.7735139202407826,
+      "grad_norm": 0.238205224275589,
+      "learning_rate": 2.9643224293745954e-05,
+      "loss": 0.7637,
+      "step": 2570
+    },
+    {
+      "epoch": 0.7765237020316027,
+      "grad_norm": 0.26717284321784973,
+      "learning_rate": 2.8900217449447074e-05,
+      "loss": 0.7326,
+      "step": 2580
+    },
+    {
+      "epoch": 0.7795334838224228,
+      "grad_norm": 0.2582661509513855,
+      "learning_rate": 2.8165064990227252e-05,
+      "loss": 0.7571,
+      "step": 2590
+    },
+    {
+      "epoch": 0.782543265613243,
+      "grad_norm": 0.22693775594234467,
+      "learning_rate": 2.7437848128296982e-05,
+      "loss": 0.75,
+      "step": 2600
+    },
+    {
+      "epoch": 0.7855530474040632,
+      "grad_norm": 0.21676841378211975,
+      "learning_rate": 2.6718647199222214e-05,
+      "loss": 0.7693,
+      "step": 2610
+    },
+    {
+      "epoch": 0.7885628291948834,
+      "grad_norm": 0.21357332170009613,
+      "learning_rate": 2.600754165304966e-05,
+      "loss": 0.7499,
+      "step": 2620
+    },
+    {
+      "epoch": 0.7915726109857035,
+      "grad_norm": 0.2448240965604782,
+      "learning_rate": 2.530461004553001e-05,
+      "loss": 0.7565,
+      "step": 2630
+    },
+    {
+      "epoch": 0.7945823927765236,
+      "grad_norm": 0.16155965626239777,
+      "learning_rate": 2.460993002943983e-05,
+      "loss": 0.738,
+      "step": 2640
+    },
+    {
+      "epoch": 0.7975921745673439,
+      "grad_norm": 0.22366459667682648,
+      "learning_rate": 2.3923578346003363e-05,
+      "loss": 0.746,
+      "step": 2650
+    },
+    {
+      "epoch": 0.800601956358164,
+      "grad_norm": 0.21000875532627106,
+      "learning_rate": 2.32456308164148e-05,
+      "loss": 0.7615,
+      "step": 2660
+    },
+    {
+      "epoch": 0.8036117381489842,
+      "grad_norm": 0.23698626458644867,
+      "learning_rate": 2.2576162333462402e-05,
+      "loss": 0.8101,
+      "step": 2670
+    },
+    {
+      "epoch": 0.8066215199398044,
+      "grad_norm": 0.17663483321666718,
+      "learning_rate": 2.191524685325512e-05,
+      "loss": 0.7551,
+      "step": 2680
+    },
+    {
+      "epoch": 0.8096313017306245,
+      "grad_norm": 0.193350687623024,
+      "learning_rate": 2.126295738705262e-05,
+      "loss": 0.7621,
+      "step": 2690
+    },
+    {
+      "epoch": 0.8126410835214447,
+      "grad_norm": 0.2506263256072998,
+      "learning_rate": 2.0619365993199747e-05,
+      "loss": 0.7789,
+      "step": 2700
+    },
+    {
+      "epoch": 0.8156508653122648,
+      "grad_norm": 0.23012906312942505,
+      "learning_rate": 1.9984543769166265e-05,
+      "loss": 0.7246,
+      "step": 2710
+    },
+    {
+      "epoch": 0.8186606471030851,
+      "grad_norm": 0.18312327563762665,
+      "learning_rate": 1.9358560843692787e-05,
+      "loss": 0.7295,
+      "step": 2720
+    },
+    {
+      "epoch": 0.8216704288939052,
+      "grad_norm": 0.22966724634170532,
+      "learning_rate": 1.8741486369043505e-05,
+      "loss": 0.7851,
+      "step": 2730
+    },
+    {
+      "epoch": 0.8246802106847254,
+      "grad_norm": 0.160865917801857,
+      "learning_rate": 1.8133388513367078e-05,
+      "loss": 0.7278,
+      "step": 2740
+    },
+    {
+      "epoch": 0.8276899924755455,
+      "grad_norm": 0.26794126629829407,
+      "learning_rate": 1.7534334453166068e-05,
+      "loss": 0.7247,
+      "step": 2750
+    },
+    {
+      "epoch": 0.8306997742663657,
+      "grad_norm": 0.23241771757602692,
+      "learning_rate": 1.6944390365875952e-05,
+      "loss": 0.7741,
+      "step": 2760
+    },
+    {
+      "epoch": 0.8337095560571859,
+      "grad_norm": 0.20465601980686188,
+      "learning_rate": 1.6363621422554476e-05,
+      "loss": 0.7267,
+      "step": 2770
+    },
+    {
+      "epoch": 0.836719337848006,
+      "grad_norm": 0.21765820682048798,
+      "learning_rate": 1.579209178068234e-05,
+      "loss": 0.7566,
+      "step": 2780
+    },
+    {
+      "epoch": 0.8397291196388262,
+      "grad_norm": 0.21495480835437775,
+      "learning_rate": 1.5229864577075547e-05,
+      "loss": 0.8005,
+      "step": 2790
+    },
+    {
+      "epoch": 0.8427389014296464,
+      "grad_norm": 0.19566848874092102,
+      "learning_rate": 1.4677001920910827e-05,
+      "loss": 0.7494,
+      "step": 2800
+    },
+    {
+      "epoch": 0.8457486832204665,
+      "grad_norm": 0.1817813217639923,
+      "learning_rate": 1.4133564886864381e-05,
+      "loss": 0.7289,
+      "step": 2810
+    },
+    {
+      "epoch": 0.8487584650112867,
+      "grad_norm": 0.2588653266429901,
+      "learning_rate": 1.3599613508364984e-05,
+      "loss": 0.7493,
+      "step": 2820
+    },
+    {
+      "epoch": 0.8517682468021068,
+      "grad_norm": 0.1954454928636551,
+      "learning_rate": 1.307520677096209e-05,
+      "loss": 0.7504,
+      "step": 2830
+    },
+    {
+      "epoch": 0.8547780285929271,
+      "grad_norm": 0.17466256022453308,
+      "learning_rate": 1.2560402605809707e-05,
+      "loss": 0.7705,
+      "step": 2840
+    },
+    {
+      "epoch": 0.8577878103837472,
+      "grad_norm": 0.2213626205921173,
+      "learning_rate": 1.2055257883266791e-05,
+      "loss": 0.7307,
+      "step": 2850
+    },
+    {
+      "epoch": 0.8607975921745673,
+      "grad_norm": 0.19768081605434418,
+      "learning_rate": 1.1559828406614714e-05,
+      "loss": 0.7741,
+      "step": 2860
+    },
+    {
+      "epoch": 0.8638073739653875,
+      "grad_norm": 0.22028213739395142,
+      "learning_rate": 1.1074168905892702e-05,
+      "loss": 0.7238,
+      "step": 2870
+    },
+    {
+      "epoch": 0.8668171557562077,
+      "grad_norm": 0.2675631046295166,
+      "learning_rate": 1.0598333031851881e-05,
+      "loss": 0.7315,
+      "step": 2880
+    },
+    {
+      "epoch": 0.8698269375470279,
+      "grad_norm": 0.20498494803905487,
+      "learning_rate": 1.0132373350028313e-05,
+      "loss": 0.751,
+      "step": 2890
+    },
+    {
+      "epoch": 0.872836719337848,
+      "grad_norm": 0.22774334251880646,
+      "learning_rate": 9.676341334936346e-06,
+      "loss": 0.72,
+      "step": 2900
+    },
+    {
+      "epoch": 0.8758465011286681,
+      "grad_norm": 0.18757307529449463,
+      "learning_rate": 9.230287364382007e-06,
+      "loss": 0.7433,
+      "step": 2910
+    },
+    {
+      "epoch": 0.8788562829194884,
+      "grad_norm": 0.23557019233703613,
+      "learning_rate": 8.794260713897862e-06,
+      "loss": 0.7473,
+      "step": 2920
+    },
+    {
+      "epoch": 0.8818660647103085,
+      "grad_norm": 0.24945016205310822,
+      "learning_rate": 8.368309551299536e-06,
+      "loss": 0.746,
+      "step": 2930
+    },
+    {
+      "epoch": 0.8848758465011287,
+      "grad_norm": 0.5168988704681396,
+      "learning_rate": 7.952480931364658e-06,
+      "loss": 0.7523,
+      "step": 2940
+    },
+    {
+      "epoch": 0.8878856282919488,
+      "grad_norm": 0.21044060587882996,
+      "learning_rate": 7.546820790634646e-06,
+      "loss": 0.7359,
+      "step": 2950
+    },
+    {
+      "epoch": 0.890895410082769,
+      "grad_norm": 0.2304990291595459,
+      "learning_rate": 7.1513739423402e-06,
+      "loss": 0.7201,
+      "step": 2960
+    },
+    {
+      "epoch": 0.8939051918735892,
+      "grad_norm": 0.5010607242584229,
+      "learning_rate": 6.766184071450721e-06,
+      "loss": 0.7619,
+      "step": 2970
+    },
+    {
+      "epoch": 0.8969149736644093,
+      "grad_norm": 0.1829194277524948,
+      "learning_rate": 6.391293729848435e-06,
+      "loss": 0.7654,
+      "step": 2980
+    },
+    {
+      "epoch": 0.8999247554552295,
+      "grad_norm": 0.4193117320537567,
+      "learning_rate": 6.026744331627731e-06,
+      "loss": 0.7416,
+      "step": 2990
+    },
+    {
+      "epoch": 0.9029345372460497,
+      "grad_norm": 0.2364160269498825,
+      "learning_rate": 5.672576148520137e-06,
+      "loss": 0.7516,
+      "step": 3000
+    },
+    {
+      "epoch": 0.9029345372460497,
+      "eval_loss": 0.7987983226776123,
+      "eval_runtime": 142.8043,
+      "eval_samples_per_second": 39.187,
+      "eval_steps_per_second": 4.902,
+      "step": 3000
+    },
+    {
+      "epoch": 0.9059443190368698,
+      "grad_norm": 0.1936904937028885,
+      "learning_rate": 5.328828305445477e-06,
+      "loss": 0.7357,
+      "step": 3010
+    },
+    {
+      "epoch": 0.90895410082769,
+      "grad_norm": 0.194386288523674,
+      "learning_rate": 4.9955387761897785e-06,
+      "loss": 0.7725,
+      "step": 3020
+    },
+    {
+      "epoch": 0.9119638826185101,
+      "grad_norm": 0.24116984009742737,
+      "learning_rate": 4.672744379210336e-06,
+      "loss": 0.776,
+      "step": 3030
+    },
+    {
+      "epoch": 0.9149736644093304,
+      "grad_norm": 0.1920221745967865,
+      "learning_rate": 4.360480773568321e-06,
+      "loss": 0.7323,
+      "step": 3040
+    },
+    {
+      "epoch": 0.9179834462001505,
+      "grad_norm": 0.19343051314353943,
+      "learning_rate": 4.058782454989529e-06,
+      "loss": 0.7436,
+      "step": 3050
+    },
+    {
+      "epoch": 0.9209932279909706,
+      "grad_norm": 0.2893059253692627,
+      "learning_rate": 3.767682752053714e-06,
+      "loss": 0.7538,
+      "step": 3060
+    },
+    {
+      "epoch": 0.9240030097817908,
+      "grad_norm": 0.23565465211868286,
+      "learning_rate": 3.487213822512714e-06,
+      "loss": 0.7535,
+      "step": 3070
+    },
+    {
+      "epoch": 0.927012791572611,
+      "grad_norm": 0.2228977531194687,
+      "learning_rate": 3.2174066497380086e-06,
+      "loss": 0.7572,
+      "step": 3080
+    },
+    {
+      "epoch": 0.9300225733634312,
+      "grad_norm": 0.20145723223686218,
+      "learning_rate": 2.958291039298e-06,
+      "loss": 0.7114,
+      "step": 3090
+    },
+    {
+      "epoch": 0.9330323551542513,
+      "grad_norm": 0.2287394404411316,
+      "learning_rate": 2.709895615665392e-06,
+      "loss": 0.7656,
+      "step": 3100
+    },
+    {
+      "epoch": 0.9360421369450714,
+      "grad_norm": 0.21397553384304047,
+      "learning_rate": 2.472247819055029e-06,
+      "loss": 0.7361,
+      "step": 3110
+    },
+    {
+      "epoch": 0.9390519187358917,
+      "grad_norm": 0.22272509336471558,
+      "learning_rate": 2.2453739023926113e-06,
+      "loss": 0.7789,
+      "step": 3120
+    },
+    {
+      "epoch": 0.9420617005267118,
+      "grad_norm": 0.2435009628534317,
+      "learning_rate": 2.0292989284144915e-06,
+      "loss": 0.7756,
+      "step": 3130
+    },
+    {
+      "epoch": 0.945071482317532,
+      "grad_norm": 0.17544378340244293,
+      "learning_rate": 1.8240467668990457e-06,
+      "loss": 0.7319,
+      "step": 3140
+    },
+    {
+      "epoch": 0.9480812641083521,
+      "grad_norm": 0.19456711411476135,
+      "learning_rate": 1.6296400920297383e-06,
+      "loss": 0.7469,
+      "step": 3150
+    },
+    {
+      "epoch": 0.9510910458991723,
+      "grad_norm": 0.21780775487422943,
+      "learning_rate": 1.4461003798903695e-06,
+      "loss": 0.7587,
+      "step": 3160
+    },
+    {
+      "epoch": 0.9541008276899925,
+      "grad_norm": 0.21393659710884094,
+      "learning_rate": 1.2734479060925397e-06,
+      "loss": 0.7504,
+      "step": 3170
+    },
+    {
+      "epoch": 0.9571106094808126,
+      "grad_norm": 0.21127097308635712,
+      "learning_rate": 1.1117017435358423e-06,
+      "loss": 0.7628,
+      "step": 3180
+    },
+    {
+      "epoch": 0.9601203912716328,
+      "grad_norm": 0.19377408921718597,
+      "learning_rate": 9.608797603008812e-07,
+      "loss": 0.7195,
+      "step": 3190
+    },
+    {
+      "epoch": 0.963130173062453,
+      "grad_norm": 0.27100813388824463,
+      "learning_rate": 8.209986176753948e-07,
+      "loss": 0.7476,
+      "step": 3200
+    },
+    {
+      "epoch": 0.9661399548532731,
+      "grad_norm": 0.1997697949409485,
+      "learning_rate": 6.920737683136613e-07,
+      "loss": 0.7578,
+      "step": 3210
+    },
+    {
+      "epoch": 0.9691497366440933,
+      "grad_norm": 0.3180257976055145,
+      "learning_rate": 5.741194545294648e-07,
+      "loss": 0.7537,
+      "step": 3220
+    },
+    {
+      "epoch": 0.9721595184349134,
+      "grad_norm": 0.20285405218601227,
+      "learning_rate": 4.671487067227531e-07,
+      "loss": 0.7226,
+      "step": 3230
+    },
+    {
+      "epoch": 0.9751693002257337,
+      "grad_norm": 0.1977706104516983,
+      "learning_rate": 3.711733419401453e-07,
+      "loss": 0.7487,
+      "step": 3240
+    },
+    {
+      "epoch": 0.9781790820165538,
+      "grad_norm": 0.2490547150373459,
+      "learning_rate": 2.8620396256953117e-07,
+      "loss": 0.773,
+      "step": 3250
+    },
+    {
+      "epoch": 0.9811888638073739,
+      "grad_norm": 0.18216556310653687,
+      "learning_rate": 2.122499551688084e-07,
+      "loss": 0.7348,
+      "step": 3260
+    },
+    {
+      "epoch": 0.9841986455981941,
+      "grad_norm": 0.20423290133476257,
+      "learning_rate": 1.4931948942895624e-07,
+      "loss": 0.7116,
+      "step": 3270
+    },
+    {
+      "epoch": 0.9872084273890143,
+      "grad_norm": 0.3691667914390564,
+      "learning_rate": 9.741951727152421e-08,
+      "loss": 0.7242,
+      "step": 3280
+    },
+    {
+      "epoch": 0.9902182091798345,
+      "grad_norm": 0.1876600682735443,
+      "learning_rate": 5.655577208069085e-08,
+      "loss": 0.755,
+      "step": 3290
+    },
+    {
+      "epoch": 0.9932279909706546,
+      "grad_norm": 0.23210102319717407,
+      "learning_rate": 2.6732768069825943e-08,
+      "loss": 0.7697,
+      "step": 3300
+    },
+    {
+      "epoch": 0.9962377727614747,
+      "grad_norm": 0.18966606259346008,
+      "learning_rate": 7.953799782889349e-09,
+      "loss": 0.7166,
+      "step": 3310
+    },
+    {
+      "epoch": 0.999247554552295,
+      "grad_norm": 0.5191490650177002,
+      "learning_rate": 2.2094173039999277e-10,
+      "loss": 0.7366,
+      "step": 3320
+    },
+    {
+      "epoch": 0.999849510910459,
+      "step": 3322,
+      "total_flos": 3.156428776263385e+18,
+      "train_loss": 0.7673604141007986,
+      "train_runtime": 8289.1597,
+      "train_samples_per_second": 12.826,
+      "train_steps_per_second": 0.401
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3322,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.156428776263385e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

sharegpt/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3346aa87355d907d9852641cae58e6acc630d339028c27755e650bed33a07a2e
+size 5432