Text Generation
Transformers
Safetensors
English
llama
conversational
text-generation-inference
4-bit precision
gptq
TheBloke commited on
Commit
b16ec73
1 Parent(s): b4f806d

GPTQ model commit

Browse files
config.json CHANGED
@@ -18,25 +18,20 @@
18
  "num_key_value_heads": 8,
19
  "pad_token_id": 0,
20
  "pretraining_tp": 1,
21
- "rms_norm_eps": 1e-05,
22
- "rope_scaling": null,
23
- "rope_theta": 10000.0,
24
- "tie_word_embeddings": false,
25
- "torch_dtype": "float16",
26
- "transformers_version": "4.37.0.dev0",
27
- "use_cache": true,
28
- "vocab_size": 32002,
29
  "quantization_config": {
 
30
  "bits": 3,
31
- "group_size": 128,
 
32
  "damp_percent": 0.1,
33
  "desc_act": true,
34
- "static_groups": false,
35
- "sym": true,
36
- "true_sequential": true,
37
- "model_name_or_path": null,
38
- "model_file_base_name": "model",
39
- "quant_method": "gptq",
 
40
  "modules_in_block_to_quantize": [
41
  [
42
  "self_attn.k_proj",
@@ -53,6 +48,21 @@
53
  [
54
  "mlp.down_proj"
55
  ]
56
- ]
57
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  }
 
18
  "num_key_value_heads": 8,
19
  "pad_token_id": 0,
20
  "pretraining_tp": 1,
 
 
 
 
 
 
 
 
21
  "quantization_config": {
22
+ "batch_size": 1,
23
  "bits": 3,
24
+ "block_name_to_quantize": null,
25
+ "cache_block_outputs": true,
26
  "damp_percent": 0.1,
27
  "desc_act": true,
28
+ "exllama_config": {
29
+ "version": 1
30
+ },
31
+ "group_size": 128,
32
+ "max_input_length": null,
33
+ "model_seqlen": null,
34
+ "module_name_preceding_first_block": null,
35
  "modules_in_block_to_quantize": [
36
  [
37
  "self_attn.k_proj",
 
48
  [
49
  "mlp.down_proj"
50
  ]
51
+ ],
52
+ "pad_token_id": null,
53
+ "quant_method": "gptq",
54
+ "sym": true,
55
+ "tokenizer": null,
56
+ "true_sequential": true,
57
+ "use_cuda_fp16": false,
58
+ "use_exllama": true
59
+ },
60
+ "rms_norm_eps": 1e-05,
61
+ "rope_scaling": null,
62
+ "rope_theta": 10000.0,
63
+ "tie_word_embeddings": false,
64
+ "torch_dtype": "float16",
65
+ "transformers_version": "4.37.0.dev0",
66
+ "use_cache": true,
67
+ "vocab_size": 32002
68
  }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 32000,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.37.0.dev0"
7
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:232fda050d110068c0c3fea2c3daceab76b77cd5453958f148e304774142b076
3
- size 48264579392
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f82121d63701a147db5bbb1e10f08dcdbc2344ccfc955e151faecba27828a713
3
+ size 48264579272
quantize_config.json CHANGED
@@ -1,11 +1,8 @@
1
  {
2
- "bits": 3,
3
- "group_size": 128,
4
- "damp_percent": 0.1,
5
- "desc_act": true,
6
- "static_groups": false,
7
- "sym": true,
8
- "true_sequential": true,
9
- "model_name_or_path": null,
10
- "model_file_base_name": "model"
11
  }
 
1
  {
2
+ "bits": 3,
3
+ "group_size": 128,
4
+ "damp_percent": 0.1,
5
+ "desc_act": true,
6
+ "sym": true,
7
+ "true_sequential": true
 
 
 
8
  }