Hugging Face's logo

PrunaAI
/
distributed-optimized-gpt2-1b-vtestnet-v2-bnb-8bit-smashed

Safetensors
gpt_optimized
pruna-ai
custom_code
8-bit precision
bitsandbytes
Model card Files Community
14

Upload folder using huggingface_hub

#14
by sharpenb - opened
base: refs/heads/main
←
from: refs/pr/14
Discussion Files changed
+3
-3
Files changed (3) hide show
  1. config.json +1 -1
  2. model.safetensors +1 -1
  3. smash_config.json +1 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "/covalent/.cache/models/tmpvi7v6ncsurvugdn3",
3
  "activation_function": "gelu_new",
4
  "all_reduce_scores": {
5
  "0": "NON_PARTICIPATING",
 
1
  {
2
+ "_name_or_path": "/covalent/.cache/models/tmpuwptik_jr49cz7zt",
3
  "activation_function": "gelu_new",
4
  "all_reduce_scores": {
5
  "0": "NON_PARTICIPATING",
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:754dc9ec61ac966db65c558ff8296db13440e3ade12488320bee4a040365ed06
3
  size 1207575528
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e25078a8b38e26662193af84492513b18e530631173f9a161f22fe4c7d9c8830
3
  size 1207575528
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/tmpvi7v6ncs",
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/tmpuwptik_j",
32
  "task": "",
33
  "save_load_fn": "bitsandbytes",
34
  "save_load_fn_args": {}