Hugging Face's logo

Xenova
/
llama2.c-stories110M

Text Generation
Transformers.js
PyTorch
ONNX
Transformers
llama
Model card Files Community
Xenova HF staff commited on
Commit
627d6ed
•
1 Parent(s): 49372c3

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. config.json +3 -2
  2. generation_config.json +1 -1
  3. onnx/decoder_model_merged.onnx +1 -1
  4. onnx/decoder_model_merged_quantized.onnx +2 -2
  5. onnx/decoder_model_quantized.onnx +2 -2
  6. onnx/decoder_with_past_model_quantized.onnx +2 -2
  7. quantize_config.json +67 -67
  8. tokenizer_config.json +2 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "../llama2.c-stories110M",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -17,8 +17,9 @@
17
  "pretraining_tp": 1,
18
  "rms_norm_eps": 1e-05,
19
  "rope_scaling": null,
 
20
  "tie_word_embeddings": false,
21
- "transformers_version": "4.32.0.dev0",
22
  "use_cache": true,
23
  "vocab_size": 32000
24
  }
 
1
  {
2
+ "_name_or_path": "Xenova/llama2.c-stories110M",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
17
  "pretraining_tp": 1,
18
  "rms_norm_eps": 1e-05,
19
  "rope_scaling": null,
20
+ "rope_theta": 10000.0,
21
  "tie_word_embeddings": false,
22
+ "transformers_version": "4.33.0.dev0",
23
  "use_cache": true,
24
  "vocab_size": 32000
25
  }
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
- "transformers_version": "4.32.0.dev0"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
+ "transformers_version": "4.33.0.dev0"
6
  }
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:868aa85e2ef1c5021cbf43c57a4bab855047f917cf73de5861a27c2992c66fd9
3
  size 538320735
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c076e620480ce241d3a15e5271e6f7852d804190b0d52673449d5b9cf80052d7
3
  size 538320735
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2526b5fbf36a866575131e95cf2c0a75fe7a03948e32eba5a05c25cdc1814dc2
3
- size 137148580
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33696a9f854ce32f34b0ec1bf033379514bf427258b6747eafb2da6e1798ccf2
3
+ size 136511989
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8e17c1ec579472eab18b12e182816011631dc488f1f03b4da1318300b0e4e43b
3
- size 135970716
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:969bc1893a6b71eeaaea45821666ff947e6260c867cb1da06e47cb6728f28948
3
+ size 135334125
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aff8a3d2b8d73fbadbe90c4e86c575a0560b6b9b6fd664a163abde6622aa5209
3
- size 135979037
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fef387f70bb057d890c11585f1950c9bd5c633e60afa0ce380b1389cca267a9
3
+ size 135342446
quantize_config.json CHANGED
@@ -1,105 +1,105 @@
1
  {
2
- "per_channel": true,
3
- "reduce_range": true,
4
  "per_model_config": {
5
- "decoder_model_merged": {
6
  "op_types": [
7
- "Slice",
8
- "Sqrt",
9
- "ConstantOfShape",
10
- "Softmax",
11
- "Sigmoid",
12
- "Equal",
13
- "Pow",
14
  "Unsqueeze",
15
- "Concat",
16
  "Sub",
17
- "Div",
18
- "Constant",
19
- "Expand",
20
- "Cast",
21
- "Range",
22
- "Reshape",
23
  "If",
24
- "Where",
25
  "Transpose",
26
- "Identity",
 
 
 
 
27
  "Mul",
28
- "Shape",
 
 
29
  "Add",
30
- "ReduceMean",
 
 
 
31
  "MatMul",
32
- "Less",
33
  "Gather",
34
- "Neg",
35
- "Squeeze"
 
 
 
 
36
  ],
37
  "weight_type": "QInt8"
38
  },
39
- "decoder_with_past_model": {
40
  "op_types": [
41
- "Slice",
42
- "Sqrt",
43
- "Squeeze",
44
- "Softmax",
45
- "Sigmoid",
46
- "Equal",
47
- "Pow",
48
  "Unsqueeze",
49
- "Concat",
50
  "Sub",
51
- "Div",
52
- "Constant",
53
- "Expand",
54
- "Cast",
55
- "Range",
56
- "Reshape",
57
- "Where",
58
  "If",
 
59
  "Transpose",
60
- "Identity",
 
 
 
 
61
  "Mul",
62
- "Shape",
 
 
63
  "Add",
64
- "ReduceMean",
 
 
 
65
  "MatMul",
 
66
  "Gather",
67
- "Neg",
68
- "ConstantOfShape"
 
 
 
 
69
  ],
70
  "weight_type": "QInt8"
71
  },
72
- "decoder_model": {
73
  "op_types": [
74
- "Slice",
75
- "Sqrt",
76
- "ConstantOfShape",
77
- "Softmax",
78
- "Sigmoid",
79
- "Equal",
80
- "Pow",
81
  "Unsqueeze",
82
- "Concat",
83
  "Sub",
84
- "Div",
85
- "Constant",
86
- "Expand",
87
- "Range",
88
- "Cast",
89
- "Reshape",
90
- "Where",
91
  "If",
 
92
  "Transpose",
93
- "Identity",
 
 
 
94
  "Mul",
95
- "Shape",
 
 
96
  "Add",
97
- "ReduceMean",
 
 
 
98
  "MatMul",
99
- "Less",
100
  "Gather",
101
- "Neg",
102
- "Squeeze"
 
 
 
 
103
  ],
104
  "weight_type": "QInt8"
105
  }
 
1
  {
2
+ "per_channel": false,
3
+ "reduce_range": false,
4
  "per_model_config": {
5
+ "decoder_model": {
6
  "op_types": [
 
 
 
 
 
 
 
7
  "Unsqueeze",
 
8
  "Sub",
9
+ "Sigmoid",
 
 
 
 
 
10
  "If",
11
+ "Shape",
12
  "Transpose",
13
+ "Slice",
14
+ "ReduceMean",
15
+ "Concat",
16
+ "Reshape",
17
+ "Range",
18
  "Mul",
19
+ "Identity",
20
+ "Softmax",
21
+ "Pow",
22
  "Add",
23
+ "Div",
24
+ "Equal",
25
+ "Neg",
26
+ "Constant",
27
  "MatMul",
28
+ "ConstantOfShape",
29
  "Gather",
30
+ "Where",
31
+ "Squeeze",
32
+ "Cast",
33
+ "Less",
34
+ "Expand",
35
+ "Sqrt"
36
  ],
37
  "weight_type": "QInt8"
38
  },
39
+ "decoder_model_merged": {
40
  "op_types": [
 
 
 
 
 
 
 
41
  "Unsqueeze",
 
42
  "Sub",
43
+ "Sigmoid",
 
 
 
 
 
 
44
  "If",
45
+ "Shape",
46
  "Transpose",
47
+ "Slice",
48
+ "ReduceMean",
49
+ "Concat",
50
+ "Reshape",
51
+ "Range",
52
  "Mul",
53
+ "Identity",
54
+ "Softmax",
55
+ "Pow",
56
  "Add",
57
+ "Div",
58
+ "Equal",
59
+ "Neg",
60
+ "Constant",
61
  "MatMul",
62
+ "ConstantOfShape",
63
  "Gather",
64
+ "Where",
65
+ "Squeeze",
66
+ "Cast",
67
+ "Less",
68
+ "Expand",
69
+ "Sqrt"
70
  ],
71
  "weight_type": "QInt8"
72
  },
73
+ "decoder_with_past_model": {
74
  "op_types": [
 
 
 
 
 
 
 
75
  "Unsqueeze",
 
76
  "Sub",
77
+ "Sigmoid",
 
 
 
 
 
 
78
  "If",
79
+ "Shape",
80
  "Transpose",
81
+ "Slice",
82
+ "ReduceMean",
83
+ "Concat",
84
+ "Reshape",
85
  "Mul",
86
+ "Identity",
87
+ "Softmax",
88
+ "Pow",
89
  "Add",
90
+ "Div",
91
+ "Equal",
92
+ "Neg",
93
+ "Constant",
94
  "MatMul",
95
+ "ConstantOfShape",
96
  "Gather",
97
+ "Where",
98
+ "Squeeze",
99
+ "Cast",
100
+ "Range",
101
+ "Expand",
102
+ "Sqrt"
103
  ],
104
  "weight_type": "QInt8"
105
  }
tokenizer_config.json CHANGED
@@ -27,5 +27,6 @@
27
  "normalized": true,
28
  "rstrip": false,
29
  "single_word": false
30
- }
 
31
  }
 
27
  "normalized": true,
28
  "rstrip": false,
29
  "single_word": false
30
+ },
31
+ "use_default_system_prompt": true
32
  }