sharpenb commited on
Commit
d7038bf
·
verified ·
1 Parent(s): e28066c

Upload folder using huggingface_hub (#3)

Browse files

- 9f60efa65c8a25943e7fec7604239136c60fb3bd364778272e2ddb71cf459775 (5ac27bf0382d97ff66ed2a78a0189958c8548281)

Files changed (2) hide show
  1. config.json +7 -1
  2. smash_config.json +1 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/covalent/.cache/models/tmpe_slx4rypnoxkbdh",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -51,6 +51,12 @@
51
  "rope_theta": 500000.0,
52
  "tie_word_embeddings": true,
53
  "torch_dtype": "float16",
 
 
 
 
 
 
54
  "transformers_version": "4.46.2",
55
  "use_cache": true,
56
  "vocab_size": 128256,
 
1
  {
2
+ "_name_or_path": "/covalent/.cache/models/tmpi486xracp_oaqqyd",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
51
  "rope_theta": 500000.0,
52
  "tie_word_embeddings": true,
53
  "torch_dtype": "float16",
54
+ "transformers.js_config": {
55
+ "kv_cache_dtype": {
56
+ "fp16": "float16",
57
+ "q4f16": "float16"
58
+ }
59
+ },
60
  "transformers_version": "4.46.2",
61
  "use_cache": true,
62
  "vocab_size": 128256,
smash_config.json CHANGED
@@ -28,7 +28,7 @@
28
  "quant_llm-int8_weight_bits": 8,
29
  "max_batch_size": 1,
30
  "device": "cuda",
31
- "cache_dir": "/covalent/.cache/models/tmpe_slx4ry",
32
  "task": "",
33
  "save_load_fn": "bitsandbytes",
34
  "save_load_fn_args": {}
 
28
  "quant_llm-int8_weight_bits": 8,
29
  "max_batch_size": 1,
30
  "device": "cuda",
31
+ "cache_dir": "/covalent/.cache/models/tmpi486xrac",
32
  "task": "",
33
  "save_load_fn": "bitsandbytes",
34
  "save_load_fn_args": {}