sharpenb commited on
Commit
14e9c08
·
verified ·
1 Parent(s): c99b942

Upload folder using huggingface_hub (#5)

Browse files

- 76b6c647c4c2fff8e93f4490283f0d13ffc8dd9a067f661e01ebb44c50f4acc3 (4c7c49b8f8e0064b5662cd9ec6c29e965c0e8ace)

Files changed (2) hide show
  1. config.json +6 -1
  2. smash_config.json +1 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/tmp/models/tmpadsvgufkzpb0zlb1",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -43,9 +43,14 @@
43
  "tie_word_embeddings": true,
44
  "torch_dtype": "bfloat16",
45
  "transformers.js_config": {
 
46
  "kv_cache_dtype": {
47
  "fp16": "float16",
48
  "q4f16": "float16"
 
 
 
 
49
  }
50
  },
51
  "transformers_version": "4.48.2",
 
1
  {
2
+ "_name_or_path": "/tmp/models/tmpu0f8olx6jfurqg9s",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
43
  "tie_word_embeddings": true,
44
  "torch_dtype": "bfloat16",
45
  "transformers.js_config": {
46
+ "dtype": "q4",
47
  "kv_cache_dtype": {
48
  "fp16": "float16",
49
  "q4f16": "float16"
50
+ },
51
+ "use_external_data_format": {
52
+ "model.onnx": true,
53
+ "model_fp16.onnx": true
54
  }
55
  },
56
  "transformers_version": "4.48.2",
smash_config.json CHANGED
@@ -15,7 +15,7 @@
15
  "quant_llm-int8_weight_bits": 4,
16
  "max_batch_size": 1,
17
  "device": "cuda",
18
- "cache_dir": "/tmp/models/tmpadsvgufk",
19
  "task": "",
20
  "save_load_fn": "llm-int8",
21
  "save_load_fn_args": {},
 
15
  "quant_llm-int8_weight_bits": 4,
16
  "max_batch_size": 1,
17
  "device": "cuda",
18
+ "cache_dir": "/tmp/models/tmpu0f8olx6",
19
  "task": "",
20
  "save_load_fn": "llm-int8",
21
  "save_load_fn_args": {},