Hugging Face's logo

onnx-community
/
InternLM2-ONNX

ONNX
Model card Files
xet
Community
3

InternLM2 - 1.8B, 7B AWQ-int4 ONNX export

#1
by rajeevp - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+13
-19
Files changed (18) hide show
  1. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/genai_config.json +1 -1
  2. {internlm2-7b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/model.onnx +2 -2
  3. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/model.onnx.data +0 -0
  4. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/special_tokens_map.json +0 -0
  5. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenization_internlm2.py +0 -0
  6. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenization_internlm2_fast.py +0 -0
  7. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenizer.json +0 -0
  8. {internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenizer.model +0 -0
  9. {internlm2-7b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenizer_config.json +2 -2
  10. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/genai_config.json +1 -1
  11. {internlm2-1.8b-cpu-int4 → internlm2-7b-cpu-int4-awq}/model.onnx +2 -2
  12. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/model.onnx.data +0 -0
  13. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/special_tokens_map.json +0 -0
  14. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenization_internlm2.py +0 -0
  15. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenization_internlm2_fast.py +0 -0
  16. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenizer.json +0 -0
  17. {internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenizer.model +0 -0
  18. {internlm2-1.8b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenizer_config.json +3 -9
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/genai_config.json RENAMED
@@ -27,7 +27,7 @@
27
  },
28
  "eos_token_id": 2,
29
  "pad_token_id": 2,
30
- "type": "internlm2",
31
  "vocab_size": 92544
32
  },
33
  "search": {
 
27
  },
28
  "eos_token_id": 2,
29
  "pad_token_id": 2,
30
+ "type": "llama",
31
  "vocab_size": 92544
32
  },
33
  "search": {
{internlm2-7b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/model.onnx RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7c952a2498956d1b1580f0687ffc8d56ea103ef3e11b13dcecb53632c7652de
3
- size 240378
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90f55a90d9e73491e9aa0dbb88098a20b8def5f9c6c129f9b2effb42f9f0fac0
3
+ size 179593
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/model.onnx.data RENAMED
File without changes
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/special_tokens_map.json RENAMED
File without changes
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenization_internlm2.py RENAMED
File without changes
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenization_internlm2_fast.py RENAMED
File without changes
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenizer.json RENAMED
File without changes
{internlm2-1.8b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenizer.model RENAMED
File without changes
{internlm2-7b-cpu-int4 → internlm2-1.8b-cpu-int4-awq}/tokenizer_config.json RENAMED
@@ -38,9 +38,9 @@
38
  "decode_with_prefix_space": false,
39
  "eos_token": "</s>",
40
  "extra_special_tokens": {},
41
- "model_max_length": 32768,
42
  "pad_token": "</s>",
43
  "sp_model_kwargs": null,
44
  "tokenizer_class": "InternLM2Tokenizer",
45
  "unk_token": "<unk>"
46
- }
 
38
  "decode_with_prefix_space": false,
39
  "eos_token": "</s>",
40
  "extra_special_tokens": {},
41
+ "model_max_length": 1000000000000000019884624838656,
42
  "pad_token": "</s>",
43
  "sp_model_kwargs": null,
44
  "tokenizer_class": "InternLM2Tokenizer",
45
  "unk_token": "<unk>"
46
+ }
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/genai_config.json RENAMED
@@ -27,7 +27,7 @@
27
  },
28
  "eos_token_id": 2,
29
  "pad_token_id": 2,
30
- "type": "internlm2",
31
  "vocab_size": 92544
32
  },
33
  "search": {
 
27
  },
28
  "eos_token_id": 2,
29
  "pad_token_id": 2,
30
+ "type": "llama",
31
  "vocab_size": 92544
32
  },
33
  "search": {
{internlm2-1.8b-cpu-int4 → internlm2-7b-cpu-int4-awq}/model.onnx RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2659c5468b3da32b1cd56910e1b3fca6ba2de0a04229fc3efd531a32d3abaea3
3
- size 180164
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:862b2f22bc845237107303a06832042c9d4641fab30a740b7ef6dfed99b146c8
3
+ size 239348
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/model.onnx.data RENAMED
File without changes
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/special_tokens_map.json RENAMED
File without changes
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenization_internlm2.py RENAMED
File without changes
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenization_internlm2_fast.py RENAMED
File without changes
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenizer.json RENAMED
File without changes
{internlm2-7b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenizer.model RENAMED
File without changes
{internlm2-1.8b-cpu-int4 → internlm2-7b-cpu-int4-awq}/tokenizer_config.json RENAMED
@@ -27,20 +27,14 @@
27
  "special": true
28
  }
29
  },
30
- "auto_map": {
31
- "AutoTokenizer": [
32
- "tokenization_internlm2.InternLM2Tokenizer",
33
- "tokenization_internlm2_fast.InternLM2TokenizerFast"
34
- ]
35
- },
36
  "bos_token": "<s>",
37
  "clean_up_tokenization_spaces": false,
38
  "decode_with_prefix_space": false,
39
  "eos_token": "</s>",
40
  "extra_special_tokens": {},
41
- "model_max_length": 32768,
42
  "pad_token": "</s>",
43
  "sp_model_kwargs": null,
44
- "tokenizer_class": "InternLM2Tokenizer",
45
  "unk_token": "<unk>"
46
- }
 
27
  "special": true
28
  }
29
  },
 
 
 
 
 
 
30
  "bos_token": "<s>",
31
  "clean_up_tokenization_spaces": false,
32
  "decode_with_prefix_space": false,
33
  "eos_token": "</s>",
34
  "extra_special_tokens": {},
35
+ "model_max_length": 1000000000000000019884624838656,
36
  "pad_token": "</s>",
37
  "sp_model_kwargs": null,
38
+ "tokenizer_class": "LlamaTokenizer",
39
  "unk_token": "<unk>"
40
+ }