metascroy commited on
Commit
3570d86
·
verified ·
1 Parent(s): 736f9a9

Upload Phi3ForCausalLM

Browse files
Files changed (2) hide show
  1. config.json +70 -20
  2. pytorch_model.bin +3 -0
config.json CHANGED
@@ -29,32 +29,82 @@
29
  "pad_token_id": 199999,
30
  "partial_rotary_factor": 0.75,
31
  "quantization_config": {
32
- "modules_to_not_convert": null,
 
33
  "quant_method": "torchao",
34
  "quant_type": {
35
  "default": {
36
  "_data": {
37
- "act_mapping_type": {
38
- "_data": "ASYMMETRIC",
39
- "_type": "MappingType"
40
- },
41
- "group_size": 32,
42
- "layout": {
43
- "_data": {},
44
- "_type": "PlainLayout",
45
- "_version": 1
46
- },
47
- "mapping_type": {
48
- "_data": "SYMMETRIC",
49
- "_type": "MappingType"
50
- },
51
- "set_inductor_config": true
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52
  },
53
- "_type": "Int8DynamicActivationInt4WeightConfig",
54
  "_version": 1
55
  }
56
  },
57
- "quant_type_kwargs": {}
 
58
  },
59
  "resid_pdrop": 0.0,
60
  "rms_norm_eps": 1e-05,
@@ -163,8 +213,8 @@
163
  },
164
  "rope_theta": 10000.0,
165
  "sliding_window": 262144,
166
- "tie_word_embeddings": true,
167
- "torch_dtype": "bfloat16",
168
  "transformers_version": "4.52.0.dev0",
169
  "use_cache": true,
170
  "vocab_size": 200064
 
29
  "pad_token_id": 199999,
30
  "partial_rotary_factor": 0.75,
31
  "quantization_config": {
32
+ "include_embedding": true,
33
+ "modules_to_not_convert": [],
34
  "quant_method": "torchao",
35
  "quant_type": {
36
  "default": {
37
  "_data": {
38
+ "module_fqn_to_config": {
39
+ "_default": {
40
+ "_data": {
41
+ "act_mapping_type": {
42
+ "_data": "ASYMMETRIC",
43
+ "_type": "MappingType"
44
+ },
45
+ "layout": {
46
+ "_data": {},
47
+ "_type": "QDQLayout",
48
+ "_version": 1
49
+ },
50
+ "weight_dtype": {
51
+ "_data": "int4",
52
+ "_type": "torch.dtype"
53
+ },
54
+ "weight_granularity": {
55
+ "_data": {
56
+ "group_size": 32
57
+ },
58
+ "_type": "PerGroup",
59
+ "_version": 1
60
+ },
61
+ "weight_mapping_type": {
62
+ "_data": "SYMMETRIC",
63
+ "_type": "MappingType"
64
+ },
65
+ "weight_scale_dtype": {
66
+ "_data": "bfloat16",
67
+ "_type": "torch.dtype"
68
+ }
69
+ },
70
+ "_type": "Int8DynamicActivationIntxWeightConfig",
71
+ "_version": 1
72
+ },
73
+ "model.embed_tokens": {
74
+ "_data": {
75
+ "granularity": {
76
+ "_data": {
77
+ "axis": 0
78
+ },
79
+ "_type": "PerAxis",
80
+ "_version": 1
81
+ },
82
+ "layout": {
83
+ "_data": {},
84
+ "_type": "QDQLayout",
85
+ "_version": 1
86
+ },
87
+ "mapping_type": {
88
+ "_data": "SYMMETRIC",
89
+ "_type": "MappingType"
90
+ },
91
+ "scale_dtype": null,
92
+ "weight_dtype": {
93
+ "_data": "int8",
94
+ "_type": "torch.dtype"
95
+ }
96
+ },
97
+ "_type": "IntxWeightOnlyConfig",
98
+ "_version": 1
99
+ }
100
+ }
101
  },
102
+ "_type": "AOPerModuleConfig",
103
  "_version": 1
104
  }
105
  },
106
+ "quant_type_kwargs": {},
107
+ "untie_embedding_weights": true
108
  },
109
  "resid_pdrop": 0.0,
110
  "rms_norm_eps": 1e-05,
 
213
  },
214
  "rope_theta": 10000.0,
215
  "sliding_window": 262144,
216
+ "tie_word_embeddings": false,
217
+ "torch_dtype": "float32",
218
  "transformers_version": "4.52.0.dev0",
219
  "use_cache": true,
220
  "vocab_size": 200064
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:968c41e72dfc2fb39ed55f2a85e207979a062db7064c601004fabf647c2614c6
3
+ size 4812014346