matichon commited on
Commit
efee2fb
·
verified ·
1 Parent(s): a13eb05

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. config.json +38 -4
  2. pytorch_model.bin +2 -2
config.json CHANGED
@@ -12,11 +12,45 @@
12
  "quant_type": {
13
  "default": {
14
  "_data": {
15
- "group_size": 128,
16
- "set_inductor_config": true
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
17
  },
18
- "_type": "Int8WeightOnlyConfig",
19
- "_version": 1
20
  }
21
  },
22
  "quant_type_kwargs": {},
 
12
  "quant_type": {
13
  "default": {
14
  "_data": {
15
+ "activation_dtype": {
16
+ "_data": "float8_e4m3fn",
17
+ "_type": "torch.dtype"
18
+ },
19
+ "activation_value_lb": null,
20
+ "activation_value_ub": null,
21
+ "granularity": [
22
+ {
23
+ "_data": {},
24
+ "_type": "PerRow",
25
+ "_version": 1
26
+ },
27
+ {
28
+ "_data": {},
29
+ "_type": "PerRow",
30
+ "_version": 1
31
+ }
32
+ ],
33
+ "kernel_preference": {
34
+ "_data": "AUTO",
35
+ "_type": "KernelPreference"
36
+ },
37
+ "mm_config": {
38
+ "_data": {
39
+ "emulate": false,
40
+ "pad_inner_dim": false,
41
+ "use_fast_accum": true
42
+ },
43
+ "_type": "Float8MMConfig",
44
+ "_version": 1
45
+ },
46
+ "set_inductor_config": true,
47
+ "weight_dtype": {
48
+ "_data": "float8_e4m3fn",
49
+ "_type": "torch.dtype"
50
+ }
51
  },
52
+ "_type": "Float8DynamicActivationFloat8WeightConfig",
53
+ "_version": 2
54
  }
55
  },
56
  "quant_type_kwargs": {},
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:042f21ef643cc57f62755bcbadfe69b568a060efa2343ccfbce53c4cfa69ae2c
3
- size 2585139752
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d51bf5db78147c31f2b59906514d239dd7c2e30a6aa3c987abf09bcf9a8ce57
3
+ size 2446753751