Hugging Face's logo

jiangchengchengNLP
/
L3.3-MS-Nevoria-70b-FP8

Text Generation
Transformers
Safetensors
English
llama
multimodal
conversational
text-generation-inference
modelopt
Model card Files
xet
Community
jiangchengchengNLP commited on
Commit
62494c4
·
verified ·
1 Parent(s): 787b4d1

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +27 -23
config.json CHANGED
@@ -37,28 +37,32 @@
37
  "use_cache": true,
38
  "vocab_size": 128256,
39
  "quantization_config": {
40
- "config_groups": {
41
- "group_0": {
42
- "input_activations": {
43
- "dynamic": false,
44
- "num_bits": 8,
45
- "type": "float"
46
- },
47
- "weights": {
48
- "dynamic": false,
49
- "num_bits": 8,
50
- "type": "float"
51
- }
52
- }
53
- },
54
- "ignore": [
55
- "lm_head"
56
- ],
57
- "quant_algo": "FP8",
58
- "kv_cache_scheme": "FP8",
59
- "producer": {
60
- "name": "modelopt",
61
- "version": "0.31.0"
62
- }
 
 
 
 
63
  }
64
  }
 
37
  "use_cache": true,
38
  "vocab_size": 128256,
39
  "quantization_config": {
40
+ "quantization": {
41
+ "config_groups": {
42
+ "group_0": {
43
+ "input_activations": {
44
+ "dynamic": false,
45
+ "num_bits": 8,
46
+ "type": "float"
47
+ },
48
+ "weights": {
49
+ "dynamic": false,
50
+ "num_bits": 8,
51
+ "type": "float"
52
+ }
53
+ }
54
+ },
55
+ "ignore": [
56
+ "lm_head"
57
+ ],
58
+ "quant_method": "modelopt",
59
+ "quant_algo": "FP8",
60
+ "kv_cache_scheme": "FP8",
61
+ "producer": {
62
+ "name": "modelopt",
63
+ "version": "0.31.0"
64
+ }
65
+ },
66
+ "quant_method": "modelopt"
67
  }
68
  }