stefan-it commited on
Commit
6b90980
·
verified ·
1 Parent(s): 59bc878

feat: add model config

Browse files
Files changed (1) hide show
  1. config.json +32 -0
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "NanoChatForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 65527,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 65531,
10
+ "final_logit_softcapping": 15.0,
11
+ "hidden_act": "relu2",
12
+ "hidden_size": 1280,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 5120,
15
+ "logits_soft_cap": 15.0,
16
+ "max_position_embeddings": 2048,
17
+ "model_type": "nanochat",
18
+ "num_attention_heads": 10,
19
+ "num_hidden_layers": 20,
20
+ "num_key_value_heads": 10,
21
+ "pad_token_id": 65531,
22
+ "rms_norm_eps": 1e-06,
23
+ "rope_parameters": {
24
+ "rope_theta": 10000.0,
25
+ "rope_type": "default"
26
+ },
27
+ "rope_theta": 10000.0,
28
+ "tie_word_embeddings": false,
29
+ "transformers_version": "5.0.0.dev0",
30
+ "use_cache": true,
31
+ "vocab_size": 65536
32
+ }