karthidev commited on
Commit
9e26875
·
verified ·
1 Parent(s): 4062a33

update config file

Browse files
Files changed (1) hide show
  1. config.json +29 -15
config.json CHANGED
@@ -3,23 +3,37 @@
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
6
- "attention_probs_dropout_prob": 0.1,
7
- "bos_token_id": 0,
8
- "eos_token_id": 1,
9
- "hidden_act": "gelu",
10
- "hidden_size": 8192,
 
 
 
 
 
11
  "initializer_range": 0.02,
12
- "intermediate_size": 32768,
13
- "max_position_embeddings": 2048,
14
- "num_attention_heads": 64,
 
 
15
  "num_hidden_layers": 32,
16
- "hidden_dropout_prob": 0.1,
17
- "vocab_size": 50265,
18
- "layer_norm_eps": 1e-6,
 
 
 
 
 
 
 
 
19
  "tie_word_embeddings": false,
20
- "torch_dtype": "float16",
 
21
  "use_cache": true,
22
- "model_type": "llama",
23
- "transformers_version": "4.43.3"
24
  }
25
-
 
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 128000,
9
+ "eos_token_id": [
10
+ 128001,
11
+ 128008,
12
+ 128009
13
+ ],
14
+ "hidden_act": "silu",
15
+ "hidden_size": 4096,
16
  "initializer_range": 0.02,
17
+ "intermediate_size": 14336,
18
+ "max_position_embeddings": 131072,
19
+ "mlp_bias": false,
20
+ "model_type": "llama",
21
+ "num_attention_heads": 32,
22
  "num_hidden_layers": 32,
23
+ "num_key_value_heads": 8,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-05,
26
+ "rope_scaling": {
27
+ "factor": 8.0,
28
+ "low_freq_factor": 1.0,
29
+ "high_freq_factor": 4.0,
30
+ "original_max_position_embeddings": 8192,
31
+ "rope_type": "llama3"
32
+ },
33
+ "rope_theta": 500000.0,
34
  "tie_word_embeddings": false,
35
+ "torch_dtype": "bfloat16",
36
+ "transformers_version": "4.42.3",
37
  "use_cache": true,
38
+ "vocab_size": 128256
 
39
  }