mlabonne commited on
Commit
d435243
1 Parent(s): f10e775

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +15 -7
config.json CHANGED
@@ -1,17 +1,20 @@
1
  {
2
- "_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
3
- "architectures": [
4
  "LlamaForCausalLM"
5
  ],
6
  "attention_bias": false,
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 128000,
9
- "eos_token_id": 128009,
 
 
 
 
10
  "hidden_act": "silu",
11
  "hidden_size": 4096,
12
  "initializer_range": 0.02,
13
  "intermediate_size": 14336,
14
- "max_position_embeddings": 8192,
15
  "mlp_bias": false,
16
  "model_type": "llama",
17
  "num_attention_heads": 32,
@@ -19,11 +22,16 @@
19
  "num_key_value_heads": 8,
20
  "pretraining_tp": 1,
21
  "rms_norm_eps": 1e-05,
22
- "rope_scaling": null,
23
- "rope_theta": 500000.0,
 
 
 
 
 
24
  "tie_word_embeddings": false,
25
  "torch_dtype": "bfloat16",
26
- "transformers_version": "4.42.4",
27
  "use_cache": true,
28
  "vocab_size": 128256
29
  }
 
1
  {
2
+ = "architectures": [
 
3
  "LlamaForCausalLM"
4
  ],
5
  "attention_bias": false,
6
  "attention_dropout": 0.0,
7
  "bos_token_id": 128000,
8
+ "eos_token_id": [
9
+ 128001,
10
+ 128008,
11
+ 128009
12
+ ],
13
  "hidden_act": "silu",
14
  "hidden_size": 4096,
15
  "initializer_range": 0.02,
16
  "intermediate_size": 14336,
17
+ "max_position_embeddings": 131072,
18
  "mlp_bias": false,
19
  "model_type": "llama",
20
  "num_attention_heads": 32,
 
22
  "num_key_value_heads": 8,
23
  "pretraining_tp": 1,
24
  "rms_norm_eps": 1e-05,
25
+ "rope_scaling": {
26
+ "factor": 8.0,
27
+ "low_freq_factor": 1.0,
28
+ "high_freq_factor": 4.0,
29
+ "original_max_position_embeddings": 8192,
30
+ "rope_type": "llama3"
31
+ },
32
  "tie_word_embeddings": false,
33
  "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.42.3",
35
  "use_cache": true,
36
  "vocab_size": 128256
37
  }