stefan-it commited on
Commit
0049e29
·
verified ·
1 Parent(s): 35fae2a

feat: add model config

Browse files
Files changed (1) hide show
  1. config.json +31 -0
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "hidden_act": "relu2",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "logits_soft_cap": 15.0,
15
+ "max_position_embeddings": 2048,
16
+ "model_type": "nanochat",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 32,
19
+ "num_key_value_heads": 16,
20
+ "pad_token_id": 65531,
21
+ "rms_norm_eps": 1e-06,
22
+ "rope_parameters": {
23
+ "rope_theta": 10000.0,
24
+ "rope_type": "default"
25
+ },
26
+ "rope_theta": 10000.0,
27
+ "tie_word_embeddings": false,
28
+ "transformers_version": "5.0.0.dev0",
29
+ "use_cache": true,
30
+ "vocab_size": 65536
31
+ }