algo-dev99 commited on
Commit
8938809
·
verified ·
1 Parent(s): 84e8513

upload checkpoint staged

Browse files
config.json CHANGED
@@ -104,7 +104,7 @@
104
  "num_routed_experts": 64,
105
  "num_shared_experts": 1,
106
  "num_v_heads": null,
107
- "pad_token_id": 248044,
108
  "quasar_layers": 4,
109
  "residual_scale": 0.1,
110
  "rms_norm_eps": 1e-06,
 
104
  "num_routed_experts": 64,
105
  "num_shared_experts": 1,
106
  "num_v_heads": null,
107
+ "pad_token_id": null,
108
  "quasar_layers": 4,
109
  "residual_scale": 0.1,
110
  "rms_norm_eps": 1e-06,
generation_config.json CHANGED
@@ -1,10 +1,7 @@
1
  {
2
  "_from_model_config": true,
3
- "eos_token_id": [
4
- 248044,
5
- 2
6
- ],
7
- "pad_token_id": 248044,
8
  "transformers_version": "5.7.0",
9
  "use_cache": true
10
  }
 
1
  {
2
  "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
 
 
 
5
  "transformers_version": "5.7.0",
6
  "use_cache": true
7
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:43469b8b6e808689d714c6faa7de0d77a7fba61784aba325f729a8443f0b3d14
3
  size 5879960192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b573eb5ad75d1a717cfe87f2783b68e98c50a1a8870944e0310d2c58b1b391e
3
  size 5879960192
tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9fee140e1e6ed13c7070c891349ce980eb030618221dde35dd00e44f8cfbdacf
3
- size 19989592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
3
+ size 19989325
tokenizer_config.json CHANGED
@@ -6,11 +6,12 @@
6
  "backend": "tokenizers",
7
  "bos_token": null,
8
  "clean_up_tokenization_spaces": false,
9
- "eos_token": "<|endoftext|>",
10
  "errors": "replace",
11
  "image_token": "<|image_pad|>",
12
  "is_local": true,
13
  "local_files_only": false,
 
14
  "model_max_length": 262144,
15
  "model_specific_special_tokens": {
16
  "audio_bos_token": "<|audio_start|>",
@@ -24,7 +25,10 @@
24
  "pad_token": "<|endoftext|>",
25
  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
26
  "split_special_tokens": false,
 
27
  "tokenizer_class": "Qwen2Tokenizer",
 
 
28
  "unk_token": null,
29
  "video_token": "<|video_pad|>",
30
  "vision_bos_token": "<|vision_start|>",
 
6
  "backend": "tokenizers",
7
  "bos_token": null,
8
  "clean_up_tokenization_spaces": false,
9
+ "eos_token": "<|im_end|>",
10
  "errors": "replace",
11
  "image_token": "<|image_pad|>",
12
  "is_local": true,
13
  "local_files_only": false,
14
+ "max_length": 2048,
15
  "model_max_length": 262144,
16
  "model_specific_special_tokens": {
17
  "audio_bos_token": "<|audio_start|>",
 
25
  "pad_token": "<|endoftext|>",
26
  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
27
  "split_special_tokens": false,
28
+ "stride": 0,
29
  "tokenizer_class": "Qwen2Tokenizer",
30
+ "truncation_side": "right",
31
+ "truncation_strategy": "longest_first",
32
  "unk_token": null,
33
  "video_token": "<|video_pad|>",
34
  "vision_bos_token": "<|vision_start|>",