Hugging Face's logo

TheBloke
/
Ziya-Coding-34B-v1.0-GPTQ

Text Generation
Transformers
Safetensors
Chinese
English
llama
text-generation-inference
4-bit precision
gptq
Model card Files Community
1
TheBloke commited on
Commit
8f15b18
•
1 Parent(s): 3d9cc30

GPTQ model commit

Browse files
Files changed (3) hide show
  1. config.json +1 -1
  2. model.safetensors +2 -2
  3. quantize_config.json +1 -1
config.json CHANGED
@@ -27,7 +27,7 @@
27
  "vocab_size": 32000,
28
  "quantization_config": {
29
  "bits": 8,
30
- "group_size": 128,
31
  "damp_percent": 0.1,
32
  "desc_act": true,
33
  "sym": true,
 
27
  "vocab_size": 32000,
28
  "quantization_config": {
29
  "bits": 8,
30
+ "group_size": -1,
31
  "damp_percent": 0.1,
32
  "desc_act": true,
33
  "sym": true,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3583c44da347f3282751e8e822e52aabf97d84a9ff2125a94ef0e89a4ac0460a
3
- size 35068270776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5aa093e5dd43ac87b9eb2e79402d056fb19c1a299cf9c19b5127ec471b1e5b90
3
+ size 34299876792
quantize_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "bits": 8,
3
- "group_size": 128,
4
  "damp_percent": 0.1,
5
  "desc_act": true,
6
  "sym": true,
 
1
  {
2
  "bits": 8,
3
+ "group_size": -1,
4
  "damp_percent": 0.1,
5
  "desc_act": true,
6
  "sym": true,