Linghe-Wang commited on
Commit
b36d7d9
·
verified ·
1 Parent(s): 189e2e2

Upload model trained with Unsloth

Browse files

Upload model trained with Unsloth 2x faster

README.md CHANGED
@@ -1,6 +1,7 @@
1
  ---
2
  library_name: transformers
3
- tags: []
 
4
  ---
5
 
6
  # Model Card for Model ID
 
1
  ---
2
  library_name: transformers
3
+ tags:
4
+ - unsloth
5
  ---
6
 
7
  # Model Card for Model ID
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/workspace/results/unsloth/Llama-3.2-3B-bnb-4bit_run__LLAMA_8B_SCHOLAWRITE/model_save",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -11,6 +11,7 @@
11
  128008,
12
  128009
13
  ],
 
14
  "hidden_act": "silu",
15
  "hidden_size": 4096,
16
  "initializer_range": 0.02,
@@ -23,6 +24,18 @@
23
  "num_key_value_heads": 8,
24
  "pad_token_id": 128004,
25
  "pretraining_tp": 1,
 
 
 
 
 
 
 
 
 
 
 
 
26
  "rms_norm_eps": 1e-05,
27
  "rope_scaling": {
28
  "factor": 8.0,
@@ -33,9 +46,9 @@
33
  },
34
  "rope_theta": 500000.0,
35
  "tie_word_embeddings": false,
36
- "torch_dtype": "float32",
37
- "transformers_version": "4.44.2",
38
- "unsloth_version": "2024.10.7",
39
  "use_cache": true,
40
  "vocab_size": 128271
41
  }
 
1
  {
2
+ "_name_or_path": "/users/1/wang9257/scholawrite/results/unsloth/Llama-3.2-8B-bnb-4bit_run__LLAMA_8B_WRITING/model_save",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
11
  128008,
12
  128009
13
  ],
14
+ "head_dim": 128,
15
  "hidden_act": "silu",
16
  "hidden_size": 4096,
17
  "initializer_range": 0.02,
 
24
  "num_key_value_heads": 8,
25
  "pad_token_id": 128004,
26
  "pretraining_tp": 1,
27
+ "quantization_config": {
28
+ "bnb_4bit_compute_dtype": "bfloat16",
29
+ "bnb_4bit_quant_type": "nf4",
30
+ "bnb_4bit_use_double_quant": true,
31
+ "llm_int8_enable_fp32_cpu_offload": false,
32
+ "llm_int8_has_fp16_weight": false,
33
+ "llm_int8_skip_modules": null,
34
+ "llm_int8_threshold": 6.0,
35
+ "load_in_4bit": true,
36
+ "load_in_8bit": false,
37
+ "quant_method": "bitsandbytes"
38
+ },
39
  "rms_norm_eps": 1e-05,
40
  "rope_scaling": {
41
  "factor": 8.0,
 
46
  },
47
  "rope_theta": 500000.0,
48
  "tie_word_embeddings": false,
49
+ "torch_dtype": "bfloat16",
50
+ "transformers_version": "4.46.3",
51
+ "unsloth_version": "2024.11.9",
52
  "use_cache": true,
53
  "vocab_size": 128271
54
  }
generation_config.json CHANGED
@@ -10,5 +10,5 @@
10
  "pad_token_id": 128004,
11
  "temperature": 0.6,
12
  "top_p": 0.9,
13
- "transformers_version": "4.44.2"
14
  }
 
10
  "pad_token_id": 128004,
11
  "temperature": 0.6,
12
  "top_p": 0.9,
13
+ "transformers_version": "4.46.3"
14
  }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba3c001cdda1a98ff2ae01519bde4f7b94a08dddf3e8676ba18e67bdf807759d
3
+ size 4652196002
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f86fdd44eef52726da0aa88fd64321a42c789e0d96913c68ffd3195e2dada75
3
+ size 1050796160
model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff