JessonWong commited on
Commit
b6ce153
·
verified ·
1 Parent(s): 0453fdc

Upload LlamaForCausalLM

Browse files
Files changed (3) hide show
  1. config.json +13 -23
  2. generation_config.json +1 -1
  3. model.safetensors +2 -2
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "meta-llama/Llama-3.1-8B-Instruct",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -12,45 +12,35 @@
12
  128008,
13
  128009
14
  ],
15
- "head_dim": 128,
16
  "hidden_act": "silu",
17
- "hidden_size": 4096,
18
  "initializer_range": 0.02,
19
- "intermediate_size": 14336,
20
  "max_position_embeddings": 131072,
21
  "mlp_bias": false,
22
  "model_type": "llama",
23
  "num_attention_heads": 32,
24
- "num_hidden_layers": 32,
25
  "num_key_value_heads": 8,
 
 
 
 
 
26
  "pretraining_tp": 1,
27
- "quantization_config": {
28
- "_load_in_4bit": true,
29
- "_load_in_8bit": false,
30
- "bnb_4bit_compute_dtype": "float32",
31
- "bnb_4bit_quant_storage": "uint8",
32
- "bnb_4bit_quant_type": "fp4",
33
- "bnb_4bit_use_double_quant": false,
34
- "llm_int8_enable_fp32_cpu_offload": false,
35
- "llm_int8_has_fp16_weight": false,
36
- "llm_int8_skip_modules": null,
37
- "llm_int8_threshold": 6.0,
38
- "load_in_4bit": true,
39
- "load_in_8bit": false,
40
- "quant_method": "bitsandbytes"
41
- },
42
  "rms_norm_eps": 1e-05,
43
  "rope_scaling": {
44
- "factor": 8.0,
45
  "high_freq_factor": 4.0,
46
  "low_freq_factor": 1.0,
47
  "original_max_position_embeddings": 8192,
48
  "rope_type": "llama3"
49
  },
50
  "rope_theta": 500000.0,
51
- "tie_word_embeddings": false,
52
  "torch_dtype": "bfloat16",
53
- "transformers_version": "4.47.0",
54
  "use_cache": true,
55
  "vocab_size": 128256
56
  }
 
1
  {
2
+ "_name_or_path": "meta-llama/Llama-3.2-1B-Instruct",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
12
  128008,
13
  128009
14
  ],
15
+ "head_dim": 64,
16
  "hidden_act": "silu",
17
+ "hidden_size": 2048,
18
  "initializer_range": 0.02,
19
+ "intermediate_size": 8192,
20
  "max_position_embeddings": 131072,
21
  "mlp_bias": false,
22
  "model_type": "llama",
23
  "num_attention_heads": 32,
24
+ "num_hidden_layers": 16,
25
  "num_key_value_heads": 8,
26
+ "pad_token_id": [
27
+ 128001,
28
+ 128008,
29
+ 128009
30
+ ],
31
  "pretraining_tp": 1,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
32
  "rms_norm_eps": 1e-05,
33
  "rope_scaling": {
34
+ "factor": 32.0,
35
  "high_freq_factor": 4.0,
36
  "low_freq_factor": 1.0,
37
  "original_max_position_embeddings": 8192,
38
  "rope_type": "llama3"
39
  },
40
  "rope_theta": 500000.0,
41
+ "tie_word_embeddings": true,
42
  "torch_dtype": "bfloat16",
43
+ "transformers_version": "4.46.3",
44
  "use_cache": true,
45
  "vocab_size": 128256
46
  }
generation_config.json CHANGED
@@ -8,5 +8,5 @@
8
  ],
9
  "temperature": 0.6,
10
  "top_p": 0.9,
11
- "transformers_version": "4.47.0"
12
  }
 
8
  ],
9
  "temperature": 0.6,
10
  "top_p": 0.9,
11
+ "transformers_version": "4.46.3"
12
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1392b4ecf7c9e8669e187fcf7b4110a0ee1ba722a88e9db275b0d13a43ed5ecf
3
- size 436885840
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d4d6ba6ce62a69ac2c027c69e683efa3b061a6eac6ad5aec657ef802be47072
3
+ size 149832