Hugging Face's logo

Xenova
/
llama2.c-stories15M

Text Generation
Transformers.js
PyTorch
ONNX
Transformers
llama
text-generation-inference
Inference Endpoints
Model card Files Community
2
Xenova HF staff commited on
Commit
3a5b4ef
·
1 Parent(s): 2580f91

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 +73 -73
  8. tokenizer_config.json +2 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "Xenova/llama2.c",
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-stories15M",
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:ee3d45805b922adb525ec778db4b2692db680561df445eebc94fce85d88985e2
3
  size 98260154
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6816b265a96d80bb85a622c394dc819b269680f327b82ea64f3c77f32117d84e
3
  size 98260154
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3ab44c99230849a428c3e3f7fb2e91b05f5b4642c19a1ec5d3877f8302cd1689
3
- size 25528662
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:473abf17f5284a881ea1086cfead870211b933365a91cead05e2274f94b7d5cb
3
+ size 25279161
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8a16c83ee7a1904271ba34b81d4f8e978cfd8cef34d99245e2453aff29d90c22
3
- size 25098988
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83c62704973668509b573854d3c24d3d47322db722538f0ce6064bc9f425802e
3
+ size 24849487
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:652587ee72f73418d7816f093df492ad8005ddcf3fc970885f7658ade6a2bca8
3
- size 25098621
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dafb7a1a57c760eac3203352c6db49456532166dca7ff046bdb5b7127d3eadec
3
+ size 24849120
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
- "Unsqueeze",
 
 
 
 
 
 
 
 
 
 
8
  "Sigmoid",
9
- "Concat",
10
- "Expand",
11
  "Transpose",
12
- "Shape",
13
- "Range",
14
- "Constant",
15
  "Neg",
16
- "If",
17
- "Cast",
18
  "Softmax",
19
- "Gather",
20
- "Add",
21
- "Equal",
22
- "Reshape",
23
- "MatMul",
24
- "ConstantOfShape",
25
  "Div",
26
- "Slice",
27
- "Sub",
28
- "Mul",
29
- "Where",
30
  "Less",
31
- "Pow",
32
- "ReduceMean",
33
- "Sqrt",
34
- "Squeeze",
35
- "Identity"
36
  ],
37
  "weight_type": "QInt8"
38
  },
39
- "decoder_with_past_model": {
40
  "op_types": [
41
- "Unsqueeze",
 
 
 
 
 
 
 
 
 
 
42
  "Sigmoid",
43
- "Concat",
44
- "Expand",
45
  "Transpose",
46
- "Shape",
47
- "Range",
48
- "Constant",
49
  "Neg",
50
- "If",
51
- "Cast",
52
  "Softmax",
 
 
 
53
  "Gather",
54
- "Add",
55
- "Equal",
56
- "Reshape",
 
 
57
  "MatMul",
58
- "ConstantOfShape",
59
- "Div",
60
- "Slice",
61
- "Sub",
62
- "Mul",
63
- "Where",
64
- "Pow",
65
- "ReduceMean",
66
- "Sqrt",
67
- "Squeeze",
68
- "Identity"
69
  ],
70
  "weight_type": "QInt8"
71
  },
72
- "decoder_model": {
73
  "op_types": [
74
- "Unsqueeze",
 
 
 
 
 
 
 
 
 
 
75
  "Sigmoid",
76
- "Concat",
77
- "Expand",
78
  "Transpose",
79
- "Shape",
80
- "Range",
81
- "Constant",
82
  "Neg",
83
- "If",
84
- "Cast",
85
  "Softmax",
 
 
 
86
  "Gather",
87
- "Add",
88
- "Equal",
89
- "Reshape",
 
90
  "MatMul",
91
- "ConstantOfShape",
92
- "Div",
93
- "Slice",
94
- "Sub",
95
- "Mul",
96
- "Where",
97
- "Less",
98
- "Pow",
99
- "ReduceMean",
100
- "Sqrt",
101
- "Squeeze",
102
- "Identity"
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
+ "Squeeze",
8
+ "Identity",
9
+ "Constant",
10
+ "Add",
11
+ "Pow",
12
+ "ConstantOfShape",
13
+ "If",
14
+ "Reshape",
15
+ "Sqrt",
16
+ "Equal",
17
+ "Where",
18
  "Sigmoid",
19
+ "Sub",
20
+ "ReduceMean",
21
  "Transpose",
22
+ "Slice",
 
 
23
  "Neg",
 
 
24
  "Softmax",
25
+ "Range",
26
+ "Shape",
 
 
 
 
27
  "Div",
28
+ "Gather",
29
+ "Concat",
30
+ "Expand",
31
+ "Cast",
32
  "Less",
33
+ "Unsqueeze",
34
+ "MatMul",
35
+ "Mul"
 
 
36
  ],
37
  "weight_type": "QInt8"
38
  },
39
+ "decoder_model_merged": {
40
  "op_types": [
41
+ "Squeeze",
42
+ "Identity",
43
+ "Constant",
44
+ "Add",
45
+ "Pow",
46
+ "ConstantOfShape",
47
+ "If",
48
+ "Reshape",
49
+ "Sqrt",
50
+ "Equal",
51
+ "Where",
52
  "Sigmoid",
53
+ "Sub",
54
+ "ReduceMean",
55
  "Transpose",
56
+ "Slice",
 
 
57
  "Neg",
 
 
58
  "Softmax",
59
+ "Range",
60
+ "Shape",
61
+ "Div",
62
  "Gather",
63
+ "Concat",
64
+ "Expand",
65
+ "Cast",
66
+ "Less",
67
+ "Unsqueeze",
68
  "MatMul",
69
+ "Mul"
 
 
 
 
 
 
 
 
 
 
70
  ],
71
  "weight_type": "QInt8"
72
  },
73
+ "decoder_with_past_model": {
74
  "op_types": [
75
+ "Squeeze",
76
+ "Identity",
77
+ "Constant",
78
+ "Add",
79
+ "Pow",
80
+ "ConstantOfShape",
81
+ "If",
82
+ "Reshape",
83
+ "Sqrt",
84
+ "Where",
85
+ "Equal",
86
  "Sigmoid",
87
+ "Sub",
88
+ "ReduceMean",
89
  "Transpose",
90
+ "Slice",
 
 
91
  "Neg",
 
 
92
  "Softmax",
93
+ "Range",
94
+ "Shape",
95
+ "Div",
96
  "Gather",
97
+ "Concat",
98
+ "Expand",
99
+ "Cast",
100
+ "Unsqueeze",
101
  "MatMul",
102
+ "Mul"
 
 
 
 
 
 
 
 
 
 
 
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
  }