Hugging Face's logo

AgileGrowth
/
ai-meat-grader

Safetensors
qwen2_vl
Model card Files
xet
Community
lee1111 commited on
Commit
b99cb2f
·
verified ·
1 Parent(s): b12c383

Upload folder using huggingface_hub

Browse files
Files changed (13) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +16 -0
  3. chat_template.json +3 -0
  4. config.json +48 -0
  5. generation_config.json +14 -0
  6. merges.txt +0 -0
  7. model.safetensors +3 -0
  8. output.20241021134301.log.txt +501 -0
  9. preprocessor_config.json +29 -0
  10. special_tokens_map.json +31 -0
  11. tokenizer.json +3 -0
  12. tokenizer_config.json +146 -0
  13. vocab.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
added_tokens.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<|box_end|>": 151649,
3
+ "<|box_start|>": 151648,
4
+ "<|endoftext|>": 151643,
5
+ "<|im_end|>": 151645,
6
+ "<|im_start|>": 151644,
7
+ "<|image_pad|>": 151655,
8
+ "<|object_ref_end|>": 151647,
9
+ "<|object_ref_start|>": 151646,
10
+ "<|quad_end|>": 151651,
11
+ "<|quad_start|>": 151650,
12
+ "<|video_pad|>": 151656,
13
+ "<|vision_end|>": 151653,
14
+ "<|vision_pad|>": 151654,
15
+ "<|vision_start|>": 151652
16
+ }
chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
+ }
config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2-VL-2B-Instruct",
3
+ "architectures": [
4
+ "Qwen2VLForConditionalGeneration"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1536,
11
+ "image_token_id": 151655,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8960,
14
+ "max_position_embeddings": 32768,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen2_vl",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 2,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": {
22
+ "mrope_section": [
23
+ 16,
24
+ 24,
25
+ 24
26
+ ],
27
+ "rope_type": "default",
28
+ "type": "default"
29
+ },
30
+ "rope_theta": 1000000.0,
31
+ "sliding_window": 32768,
32
+ "tie_word_embeddings": true,
33
+ "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.45.2",
35
+ "use_cache": true,
36
+ "use_sliding_window": false,
37
+ "video_token_id": 151656,
38
+ "vision_config": {
39
+ "hidden_size": 1536,
40
+ "in_chans": 3,
41
+ "model_type": "qwen2_vl",
42
+ "spatial_patch_size": 14
43
+ },
44
+ "vision_end_token_id": 151653,
45
+ "vision_start_token_id": 151652,
46
+ "vision_token_id": 151654,
47
+ "vocab_size": 151936
48
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "attn_implementation": "flash_attention_2",
3
+ "bos_token_id": 151643,
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 151645,
7
+ 151643
8
+ ],
9
+ "pad_token_id": 151643,
10
+ "temperature": 0.01,
11
+ "top_k": 1,
12
+ "top_p": 0.001,
13
+ "transformers_version": "4.45.2"
14
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a111b34aab66a5a53e90df46918080d8137321e4c5aac8490d44260a7d07cd76
3
+ size 4418050848
output.20241021134301.log.txt ADDED
@@ -0,0 +1,501 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2024-10-21 13:43:07-finetune.py:240-INFO >> Batch 2 of epoch 1/10, average training loss of previous 2 batches: 1.383868396282196
2
+ 2024-10-21 13:43:08-finetune.py:240-INFO >> Batch 4 of epoch 1/10, average training loss of previous 2 batches: 1.4955955147743225
3
+ 2024-10-21 13:43:09-finetune.py:240-INFO >> Batch 6 of epoch 1/10, average training loss of previous 2 batches: 1.2759928703308105
4
+ 2024-10-21 13:43:10-finetune.py:240-INFO >> Batch 8 of epoch 1/10, average training loss of previous 2 batches: 1.6038963794708252
5
+ 2024-10-21 13:43:11-finetune.py:240-INFO >> Batch 10 of epoch 1/10, average training loss of previous 2 batches: 1.2483851313591003
6
+ 2024-10-21 13:43:12-finetune.py:240-INFO >> Batch 12 of epoch 1/10, average training loss of previous 2 batches: 1.3747590780258179
7
+ 2024-10-21 13:43:12-finetune.py:240-INFO >> Batch 14 of epoch 1/10, average training loss of previous 2 batches: 1.27400141954422
8
+ 2024-10-21 13:43:13-finetune.py:240-INFO >> Batch 16 of epoch 1/10, average training loss of previous 2 batches: 1.441308856010437
9
+ 2024-10-21 13:43:14-finetune.py:240-INFO >> Batch 18 of epoch 1/10, average training loss of previous 2 batches: 1.2634166479110718
10
+ 2024-10-21 13:43:15-finetune.py:240-INFO >> Batch 20 of epoch 1/10, average training loss of previous 2 batches: 1.1642013788223267
11
+ 2024-10-21 13:43:16-finetune.py:240-INFO >> Batch 22 of epoch 1/10, average training loss of previous 2 batches: 1.3199244737625122
12
+ 2024-10-21 13:43:17-finetune.py:240-INFO >> Batch 24 of epoch 1/10, average training loss of previous 2 batches: 1.2249361276626587
13
+ 2024-10-21 13:43:18-finetune.py:240-INFO >> Batch 26 of epoch 1/10, average training loss of previous 2 batches: 1.3165916800498962
14
+ 2024-10-21 13:43:18-finetune.py:240-INFO >> Batch 28 of epoch 1/10, average training loss of previous 2 batches: 1.1227564811706543
15
+ 2024-10-21 13:43:19-finetune.py:240-INFO >> Batch 30 of epoch 1/10, average training loss of previous 2 batches: 1.022961974143982
16
+ 2024-10-21 13:43:20-finetune.py:240-INFO >> Batch 32 of epoch 1/10, average training loss of previous 2 batches: 1.1248258352279663
17
+ 2024-10-21 13:43:21-finetune.py:240-INFO >> Batch 34 of epoch 1/10, average training loss of previous 2 batches: 1.1138423681259155
18
+ 2024-10-21 13:43:22-finetune.py:240-INFO >> Batch 36 of epoch 1/10, average training loss of previous 2 batches: 1.2348712086677551
19
+ 2024-10-21 13:43:23-finetune.py:240-INFO >> Batch 38 of epoch 1/10, average training loss of previous 2 batches: 1.032978355884552
20
+ 2024-10-21 13:43:23-finetune.py:240-INFO >> Batch 40 of epoch 1/10, average training loss of previous 2 batches: 1.0560821890830994
21
+ 2024-10-21 13:43:24-finetune.py:240-INFO >> Batch 42 of epoch 1/10, average training loss of previous 2 batches: 1.0707827806472778
22
+ 2024-10-21 13:43:25-finetune.py:240-INFO >> Batch 44 of epoch 1/10, average training loss of previous 2 batches: 1.0700677037239075
23
+ 2024-10-21 13:43:26-finetune.py:240-INFO >> Batch 46 of epoch 1/10, average training loss of previous 2 batches: 1.1728523969650269
24
+ 2024-10-21 13:43:27-finetune.py:240-INFO >> Batch 48 of epoch 1/10, average training loss of previous 2 batches: 1.207688808441162
25
+ 2024-10-21 13:43:28-finetune.py:240-INFO >> Batch 50 of epoch 1/10, average training loss of previous 2 batches: 0.9196622967720032
26
+ 2024-10-21 13:43:29-finetune.py:240-INFO >> Batch 52 of epoch 1/10, average training loss of previous 2 batches: 1.1481328010559082
27
+ 2024-10-21 13:43:29-finetune.py:240-INFO >> Batch 54 of epoch 1/10, average training loss of previous 2 batches: 1.104943871498108
28
+ 2024-10-21 13:43:30-finetune.py:240-INFO >> Batch 56 of epoch 1/10, average training loss of previous 2 batches: 0.9173874855041504
29
+ 2024-10-21 13:43:31-finetune.py:240-INFO >> Batch 58 of epoch 1/10, average training loss of previous 2 batches: 0.9855371713638306
30
+ 2024-10-21 13:43:32-finetune.py:240-INFO >> Batch 60 of epoch 1/10, average training loss of previous 2 batches: 1.1189665794372559
31
+ 2024-10-21 13:43:33-finetune.py:240-INFO >> Batch 62 of epoch 1/10, average training loss of previous 2 batches: 0.9940662384033203
32
+ 2024-10-21 13:43:34-finetune.py:240-INFO >> Batch 64 of epoch 1/10, average training loss of previous 2 batches: 1.120991289615631
33
+ 2024-10-21 13:43:34-finetune.py:240-INFO >> Batch 66 of epoch 1/10, average training loss of previous 2 batches: 0.9912855923175812
34
+ 2024-10-21 13:43:35-finetune.py:240-INFO >> Batch 68 of epoch 1/10, average training loss of previous 2 batches: 0.9637334048748016
35
+ 2024-10-21 13:43:36-finetune.py:240-INFO >> Batch 70 of epoch 1/10, average training loss of previous 2 batches: 1.0715060234069824
36
+ 2024-10-21 13:43:37-finetune.py:240-INFO >> Batch 72 of epoch 1/10, average training loss of previous 2 batches: 0.8575984835624695
37
+ 2024-10-21 13:43:38-finetune.py:240-INFO >> Batch 74 of epoch 1/10, average training loss of previous 2 batches: 1.012690782546997
38
+ 2024-10-21 13:43:39-finetune.py:240-INFO >> Batch 76 of epoch 1/10, average training loss of previous 2 batches: 0.9796179533004761
39
+ 2024-10-21 13:43:40-finetune.py:240-INFO >> Batch 78 of epoch 1/10, average training loss of previous 2 batches: 0.9251889288425446
40
+ 2024-10-21 13:43:40-finetune.py:240-INFO >> Batch 80 of epoch 1/10, average training loss of previous 2 batches: 1.2643179893493652
41
+ 2024-10-21 13:43:41-finetune.py:240-INFO >> Batch 82 of epoch 1/10, average training loss of previous 2 batches: 1.1511635184288025
42
+ 2024-10-21 13:43:42-finetune.py:240-INFO >> Batch 84 of epoch 1/10, average training loss of previous 2 batches: 1.0833743810653687
43
+ 2024-10-21 13:43:43-finetune.py:240-INFO >> Batch 86 of epoch 1/10, average training loss of previous 2 batches: 1.0614412426948547
44
+ 2024-10-21 13:43:44-finetune.py:240-INFO >> Batch 88 of epoch 1/10, average training loss of previous 2 batches: 1.0845468044281006
45
+ 2024-10-21 13:43:45-finetune.py:240-INFO >> Batch 90 of epoch 1/10, average training loss of previous 2 batches: 1.0898866057395935
46
+ 2024-10-21 13:43:45-finetune.py:240-INFO >> Batch 92 of epoch 1/10, average training loss of previous 2 batches: 1.106179416179657
47
+ 2024-10-21 13:43:46-finetune.py:240-INFO >> Batch 94 of epoch 1/10, average training loss of previous 2 batches: 0.9967300891876221
48
+ 2024-10-21 13:43:47-finetune.py:240-INFO >> Batch 96 of epoch 1/10, average training loss of previous 2 batches: 1.0550669431686401
49
+ 2024-10-21 13:43:48-finetune.py:240-INFO >> Batch 98 of epoch 1/10, average training loss of previous 2 batches: 1.003680169582367
50
+ 2024-10-21 13:43:49-finetune.py:240-INFO >> Batch 100 of epoch 1/10, average training loss of previous 2 batches: 1.0866395235061646
51
+ 2024-10-21 13:43:50-finetune.py:240-INFO >> Batch 2 of epoch 2/10, average training loss of previous 2 batches: 0.8128059804439545
52
+ 2024-10-21 13:43:51-finetune.py:240-INFO >> Batch 4 of epoch 2/10, average training loss of previous 2 batches: 0.9414032399654388
53
+ 2024-10-21 13:43:51-finetune.py:240-INFO >> Batch 6 of epoch 2/10, average training loss of previous 2 batches: 0.8227186799049377
54
+ 2024-10-21 13:43:52-finetune.py:240-INFO >> Batch 8 of epoch 2/10, average training loss of previous 2 batches: 1.1816160678863525
55
+ 2024-10-21 13:43:53-finetune.py:240-INFO >> Batch 10 of epoch 2/10, average training loss of previous 2 batches: 0.8436281085014343
56
+ 2024-10-21 13:43:54-finetune.py:240-INFO >> Batch 12 of epoch 2/10, average training loss of previous 2 batches: 0.9917052984237671
57
+ 2024-10-21 13:43:55-finetune.py:240-INFO >> Batch 14 of epoch 2/10, average training loss of previous 2 batches: 0.9916376769542694
58
+ 2024-10-21 13:43:56-finetune.py:240-INFO >> Batch 16 of epoch 2/10, average training loss of previous 2 batches: 1.1409175992012024
59
+ 2024-10-21 13:43:57-finetune.py:240-INFO >> Batch 18 of epoch 2/10, average training loss of previous 2 batches: 0.9628605842590332
60
+ 2024-10-21 13:43:57-finetune.py:240-INFO >> Batch 20 of epoch 2/10, average training loss of previous 2 batches: 0.9078276753425598
61
+ 2024-10-21 13:43:58-finetune.py:240-INFO >> Batch 22 of epoch 2/10, average training loss of previous 2 batches: 1.0480610132217407
62
+ 2024-10-21 13:43:59-finetune.py:240-INFO >> Batch 24 of epoch 2/10, average training loss of previous 2 batches: 0.9932570159435272
63
+ 2024-10-21 13:44:00-finetune.py:240-INFO >> Batch 26 of epoch 2/10, average training loss of previous 2 batches: 1.0300598442554474
64
+ 2024-10-21 13:44:01-finetune.py:240-INFO >> Batch 28 of epoch 2/10, average training loss of previous 2 batches: 0.8787409961223602
65
+ 2024-10-21 13:44:02-finetune.py:240-INFO >> Batch 30 of epoch 2/10, average training loss of previous 2 batches: 0.8204362094402313
66
+ 2024-10-21 13:44:03-finetune.py:240-INFO >> Batch 32 of epoch 2/10, average training loss of previous 2 batches: 0.8951331973075867
67
+ 2024-10-21 13:44:03-finetune.py:240-INFO >> Batch 34 of epoch 2/10, average training loss of previous 2 batches: 0.9430911242961884
68
+ 2024-10-21 13:44:04-finetune.py:240-INFO >> Batch 36 of epoch 2/10, average training loss of previous 2 batches: 1.037635713815689
69
+ 2024-10-21 13:44:05-finetune.py:240-INFO >> Batch 38 of epoch 2/10, average training loss of previous 2 batches: 0.8533597886562347
70
+ 2024-10-21 13:44:06-finetune.py:240-INFO >> Batch 40 of epoch 2/10, average training loss of previous 2 batches: 0.8404317200183868
71
+ 2024-10-21 13:44:07-finetune.py:240-INFO >> Batch 42 of epoch 2/10, average training loss of previous 2 batches: 0.879459410905838
72
+ 2024-10-21 13:44:08-finetune.py:240-INFO >> Batch 44 of epoch 2/10, average training loss of previous 2 batches: 0.9176307320594788
73
+ 2024-10-21 13:44:09-finetune.py:240-INFO >> Batch 46 of epoch 2/10, average training loss of previous 2 batches: 0.9854529798030853
74
+ 2024-10-21 13:44:09-finetune.py:240-INFO >> Batch 48 of epoch 2/10, average training loss of previous 2 batches: 1.0120484232902527
75
+ 2024-10-21 13:44:10-finetune.py:240-INFO >> Batch 50 of epoch 2/10, average training loss of previous 2 batches: 0.7464470267295837
76
+ 2024-10-21 13:44:11-finetune.py:240-INFO >> Batch 52 of epoch 2/10, average training loss of previous 2 batches: 0.9358537495136261
77
+ 2024-10-21 13:44:12-finetune.py:240-INFO >> Batch 54 of epoch 2/10, average training loss of previous 2 batches: 0.9208785891532898
78
+ 2024-10-21 13:44:13-finetune.py:240-INFO >> Batch 56 of epoch 2/10, average training loss of previous 2 batches: 0.7592509090900421
79
+ 2024-10-21 13:44:14-finetune.py:240-INFO >> Batch 58 of epoch 2/10, average training loss of previous 2 batches: 0.8204958140850067
80
+ 2024-10-21 13:44:15-finetune.py:240-INFO >> Batch 60 of epoch 2/10, average training loss of previous 2 batches: 0.9596169292926788
81
+ 2024-10-21 13:44:15-finetune.py:240-INFO >> Batch 62 of epoch 2/10, average training loss of previous 2 batches: 0.8344171643257141
82
+ 2024-10-21 13:44:16-finetune.py:240-INFO >> Batch 64 of epoch 2/10, average training loss of previous 2 batches: 0.979409247636795
83
+ 2024-10-21 13:44:17-finetune.py:240-INFO >> Batch 66 of epoch 2/10, average training loss of previous 2 batches: 0.8702147006988525
84
+ 2024-10-21 13:44:18-finetune.py:240-INFO >> Batch 68 of epoch 2/10, average training loss of previous 2 batches: 0.8352161645889282
85
+ 2024-10-21 13:44:19-finetune.py:240-INFO >> Batch 70 of epoch 2/10, average training loss of previous 2 batches: 0.9293053150177002
86
+ 2024-10-21 13:44:20-finetune.py:240-INFO >> Batch 72 of epoch 2/10, average training loss of previous 2 batches: 0.7029016017913818
87
+ 2024-10-21 13:44:21-finetune.py:240-INFO >> Batch 74 of epoch 2/10, average training loss of previous 2 batches: 0.8921840488910675
88
+ 2024-10-21 13:44:21-finetune.py:240-INFO >> Batch 76 of epoch 2/10, average training loss of previous 2 batches: 0.8468500375747681
89
+ 2024-10-21 13:44:22-finetune.py:240-INFO >> Batch 78 of epoch 2/10, average training loss of previous 2 batches: 0.8011757731437683
90
+ 2024-10-21 13:44:23-finetune.py:240-INFO >> Batch 80 of epoch 2/10, average training loss of previous 2 batches: 1.1114087104797363
91
+ 2024-10-21 13:44:24-finetune.py:240-INFO >> Batch 82 of epoch 2/10, average training loss of previous 2 batches: 1.0192084610462189
92
+ 2024-10-21 13:44:25-finetune.py:240-INFO >> Batch 84 of epoch 2/10, average training loss of previous 2 batches: 0.941729873418808
93
+ 2024-10-21 13:44:26-finetune.py:240-INFO >> Batch 86 of epoch 2/10, average training loss of previous 2 batches: 0.9171375632286072
94
+ 2024-10-21 13:44:27-finetune.py:240-INFO >> Batch 88 of epoch 2/10, average training loss of previous 2 batches: 0.9673053622245789
95
+ 2024-10-21 13:44:28-finetune.py:240-INFO >> Batch 90 of epoch 2/10, average training loss of previous 2 batches: 0.9618223011493683
96
+ 2024-10-21 13:44:28-finetune.py:240-INFO >> Batch 92 of epoch 2/10, average training loss of previous 2 batches: 0.9772896766662598
97
+ 2024-10-21 13:44:29-finetune.py:240-INFO >> Batch 94 of epoch 2/10, average training loss of previous 2 batches: 0.8849041163921356
98
+ 2024-10-21 13:44:30-finetune.py:240-INFO >> Batch 96 of epoch 2/10, average training loss of previous 2 batches: 0.9389393627643585
99
+ 2024-10-21 13:44:31-finetune.py:240-INFO >> Batch 98 of epoch 2/10, average training loss of previous 2 batches: 0.9033451080322266
100
+ 2024-10-21 13:44:32-finetune.py:240-INFO >> Batch 100 of epoch 2/10, average training loss of previous 2 batches: 0.970057487487793
101
+ 2024-10-21 13:44:33-finetune.py:240-INFO >> Batch 2 of epoch 3/10, average training loss of previous 2 batches: 0.736648291349411
102
+ 2024-10-21 13:44:34-finetune.py:240-INFO >> Batch 4 of epoch 3/10, average training loss of previous 2 batches: 0.8554295003414154
103
+ 2024-10-21 13:44:34-finetune.py:240-INFO >> Batch 6 of epoch 3/10, average training loss of previous 2 batches: 0.7599872350692749
104
+ 2024-10-21 13:44:35-finetune.py:240-INFO >> Batch 8 of epoch 3/10, average training loss of previous 2 batches: 1.066439151763916
105
+ 2024-10-21 13:44:36-finetune.py:240-INFO >> Batch 10 of epoch 3/10, average training loss of previous 2 batches: 0.7500385642051697
106
+ 2024-10-21 13:44:37-finetune.py:240-INFO >> Batch 12 of epoch 3/10, average training loss of previous 2 batches: 0.8974490761756897
107
+ 2024-10-21 13:44:38-finetune.py:240-INFO >> Batch 14 of epoch 3/10, average training loss of previous 2 batches: 0.9173659086227417
108
+ 2024-10-21 13:44:39-finetune.py:240-INFO >> Batch 16 of epoch 3/10, average training loss of previous 2 batches: 1.0351774096488953
109
+ 2024-10-21 13:44:40-finetune.py:240-INFO >> Batch 18 of epoch 3/10, average training loss of previous 2 batches: 0.8837870657444
110
+ 2024-10-21 13:44:41-finetune.py:240-INFO >> Batch 20 of epoch 3/10, average training loss of previous 2 batches: 0.8277487754821777
111
+ 2024-10-21 13:44:41-finetune.py:240-INFO >> Batch 22 of epoch 3/10, average training loss of previous 2 batches: 0.9443394541740417
112
+ 2024-10-21 13:44:42-finetune.py:240-INFO >> Batch 24 of epoch 3/10, average training loss of previous 2 batches: 0.9046522080898285
113
+ 2024-10-21 13:44:43-finetune.py:240-INFO >> Batch 26 of epoch 3/10, average training loss of previous 2 batches: 0.9271535575389862
114
+ 2024-10-21 13:44:44-finetune.py:240-INFO >> Batch 28 of epoch 3/10, average training loss of previous 2 batches: 0.7925187945365906
115
+ 2024-10-21 13:44:45-finetune.py:240-INFO >> Batch 30 of epoch 3/10, average training loss of previous 2 batches: 0.7540174126625061
116
+ 2024-10-21 13:44:46-finetune.py:240-INFO >> Batch 32 of epoch 3/10, average training loss of previous 2 batches: 0.8095866143703461
117
+ 2024-10-21 13:44:47-finetune.py:240-INFO >> Batch 34 of epoch 3/10, average training loss of previous 2 batches: 0.8659921884536743
118
+ 2024-10-21 13:44:47-finetune.py:240-INFO >> Batch 36 of epoch 3/10, average training loss of previous 2 batches: 0.943943202495575
119
+ 2024-10-21 13:44:48-finetune.py:240-INFO >> Batch 38 of epoch 3/10, average training loss of previous 2 batches: 0.7801710069179535
120
+ 2024-10-21 13:44:49-finetune.py:240-INFO >> Batch 40 of epoch 3/10, average training loss of previous 2 batches: 0.7536539733409882
121
+ 2024-10-21 13:44:50-finetune.py:240-INFO >> Batch 42 of epoch 3/10, average training loss of previous 2 batches: 0.7926651835441589
122
+ 2024-10-21 13:44:51-finetune.py:240-INFO >> Batch 44 of epoch 3/10, average training loss of previous 2 batches: 0.8482162654399872
123
+ 2024-10-21 13:44:52-finetune.py:240-INFO >> Batch 46 of epoch 3/10, average training loss of previous 2 batches: 0.9039164483547211
124
+ 2024-10-21 13:44:53-finetune.py:240-INFO >> Batch 48 of epoch 3/10, average training loss of previous 2 batches: 0.923221081495285
125
+ 2024-10-21 13:44:54-finetune.py:240-INFO >> Batch 50 of epoch 3/10, average training loss of previous 2 batches: 0.6753100752830505
126
+ 2024-10-21 13:44:54-finetune.py:240-INFO >> Batch 52 of epoch 3/10, average training loss of previous 2 batches: 0.8334802985191345
127
+ 2024-10-21 13:44:55-finetune.py:240-INFO >> Batch 54 of epoch 3/10, average training loss of previous 2 batches: 0.8219180703163147
128
+ 2024-10-21 13:44:56-finetune.py:240-INFO >> Batch 56 of epoch 3/10, average training loss of previous 2 batches: 0.6824038624763489
129
+ 2024-10-21 13:44:57-finetune.py:240-INFO >> Batch 58 of epoch 3/10, average training loss of previous 2 batches: 0.7407535016536713
130
+ 2024-10-21 13:44:58-finetune.py:240-INFO >> Batch 60 of epoch 3/10, average training loss of previous 2 batches: 0.8619542419910431
131
+ 2024-10-21 13:44:59-finetune.py:240-INFO >> Batch 62 of epoch 3/10, average training loss of previous 2 batches: 0.7519990801811218
132
+ 2024-10-21 13:45:00-finetune.py:240-INFO >> Batch 64 of epoch 3/10, average training loss of previous 2 batches: 0.8981192409992218
133
+ 2024-10-21 13:45:00-finetune.py:240-INFO >> Batch 66 of epoch 3/10, average training loss of previous 2 batches: 0.795328825712204
134
+ 2024-10-21 13:45:01-finetune.py:240-INFO >> Batch 68 of epoch 3/10, average training loss of previous 2 batches: 0.7598937749862671
135
+ 2024-10-21 13:45:02-finetune.py:240-INFO >> Batch 70 of epoch 3/10, average training loss of previous 2 batches: 0.8479012846946716
136
+ 2024-10-21 13:45:03-finetune.py:240-INFO >> Batch 72 of epoch 3/10, average training loss of previous 2 batches: 0.6232497692108154
137
+ 2024-10-21 13:45:04-finetune.py:240-INFO >> Batch 74 of epoch 3/10, average training loss of previous 2 batches: 0.8171462714672089
138
+ 2024-10-21 13:45:05-finetune.py:240-INFO >> Batch 76 of epoch 3/10, average training loss of previous 2 batches: 0.7621821761131287
139
+ 2024-10-21 13:45:06-finetune.py:240-INFO >> Batch 78 of epoch 3/10, average training loss of previous 2 batches: 0.7309359610080719
140
+ 2024-10-21 13:45:07-finetune.py:240-INFO >> Batch 80 of epoch 3/10, average training loss of previous 2 batches: 1.012621521949768
141
+ 2024-10-21 13:45:07-finetune.py:240-INFO >> Batch 82 of epoch 3/10, average training loss of previous 2 batches: 0.9371889233589172
142
+ 2024-10-21 13:45:08-finetune.py:240-INFO >> Batch 84 of epoch 3/10, average training loss of previous 2 batches: 0.8546919226646423
143
+ 2024-10-21 13:45:09-finetune.py:240-INFO >> Batch 86 of epoch 3/10, average training loss of previous 2 batches: 0.8201634585857391
144
+ 2024-10-21 13:45:10-finetune.py:240-INFO >> Batch 88 of epoch 3/10, average training loss of previous 2 batches: 0.8924104869365692
145
+ 2024-10-21 13:45:11-finetune.py:240-INFO >> Batch 90 of epoch 3/10, average training loss of previous 2 batches: 0.8748036623001099
146
+ 2024-10-21 13:45:12-finetune.py:240-INFO >> Batch 92 of epoch 3/10, average training loss of previous 2 batches: 0.8777934610843658
147
+ 2024-10-21 13:45:13-finetune.py:240-INFO >> Batch 94 of epoch 3/10, average training loss of previous 2 batches: 0.8174447417259216
148
+ 2024-10-21 13:45:14-finetune.py:240-INFO >> Batch 96 of epoch 3/10, average training loss of previous 2 batches: 0.8672417104244232
149
+ 2024-10-21 13:45:14-finetune.py:240-INFO >> Batch 98 of epoch 3/10, average training loss of previous 2 batches: 0.8363238275051117
150
+ 2024-10-21 13:45:15-finetune.py:240-INFO >> Batch 100 of epoch 3/10, average training loss of previous 2 batches: 0.8956538736820221
151
+ 2024-10-21 13:45:16-finetune.py:240-INFO >> Batch 2 of epoch 4/10, average training loss of previous 2 batches: 0.6835333406925201
152
+ 2024-10-21 13:45:17-finetune.py:240-INFO >> Batch 4 of epoch 4/10, average training loss of previous 2 batches: 0.7852495908737183
153
+ 2024-10-21 13:45:18-finetune.py:240-INFO >> Batch 6 of epoch 4/10, average training loss of previous 2 batches: 0.716625303030014
154
+ 2024-10-21 13:45:19-finetune.py:240-INFO >> Batch 8 of epoch 4/10, average training loss of previous 2 batches: 0.984612375497818
155
+ 2024-10-21 13:45:20-finetune.py:240-INFO >> Batch 10 of epoch 4/10, average training loss of previous 2 batches: 0.6900284886360168
156
+ 2024-10-21 13:45:21-finetune.py:240-INFO >> Batch 12 of epoch 4/10, average training loss of previous 2 batches: 0.8278656899929047
157
+ 2024-10-21 13:45:21-finetune.py:240-INFO >> Batch 14 of epoch 4/10, average training loss of previous 2 batches: 0.8600967824459076
158
+ 2024-10-21 13:45:22-finetune.py:240-INFO >> Batch 16 of epoch 4/10, average training loss of previous 2 batches: 0.9587431848049164
159
+ 2024-10-21 13:45:23-finetune.py:240-INFO >> Batch 18 of epoch 4/10, average training loss of previous 2 batches: 0.8181414604187012
160
+ 2024-10-21 13:45:24-finetune.py:240-INFO >> Batch 20 of epoch 4/10, average training loss of previous 2 batches: 0.7698231935501099
161
+ 2024-10-21 13:45:25-finetune.py:240-INFO >> Batch 22 of epoch 4/10, average training loss of previous 2 batches: 0.8695583045482635
162
+ 2024-10-21 13:45:26-finetune.py:240-INFO >> Batch 24 of epoch 4/10, average training loss of previous 2 batches: 0.8343754410743713
163
+ 2024-10-21 13:45:27-finetune.py:240-INFO >> Batch 26 of epoch 4/10, average training loss of previous 2 batches: 0.8548255264759064
164
+ 2024-10-21 13:45:28-finetune.py:240-INFO >> Batch 28 of epoch 4/10, average training loss of previous 2 batches: 0.7314669489860535
165
+ 2024-10-21 13:45:28-finetune.py:240-INFO >> Batch 30 of epoch 4/10, average training loss of previous 2 batches: 0.7021181285381317
166
+ 2024-10-21 13:45:29-finetune.py:240-INFO >> Batch 32 of epoch 4/10, average training loss of previous 2 batches: 0.7473123967647552
167
+ 2024-10-21 13:45:30-finetune.py:240-INFO >> Batch 34 of epoch 4/10, average training loss of previous 2 batches: 0.8069173097610474
168
+ 2024-10-21 13:45:31-finetune.py:240-INFO >> Batch 36 of epoch 4/10, average training loss of previous 2 batches: 0.8744145035743713
169
+ 2024-10-21 13:45:32-finetune.py:240-INFO >> Batch 38 of epoch 4/10, average training loss of previous 2 batches: 0.7218718826770782
170
+ 2024-10-21 13:45:33-finetune.py:240-INFO >> Batch 40 of epoch 4/10, average training loss of previous 2 batches: 0.6963265836238861
171
+ 2024-10-21 13:45:34-finetune.py:240-INFO >> Batch 42 of epoch 4/10, average training loss of previous 2 batches: 0.7223711013793945
172
+ 2024-10-21 13:45:34-finetune.py:240-INFO >> Batch 44 of epoch 4/10, average training loss of previous 2 batches: 0.7927519381046295
173
+ 2024-10-21 13:45:35-finetune.py:240-INFO >> Batch 46 of epoch 4/10, average training loss of previous 2 batches: 0.8368312120437622
174
+ 2024-10-21 13:45:36-finetune.py:240-INFO >> Batch 48 of epoch 4/10, average training loss of previous 2 batches: 0.8447033166885376
175
+ 2024-10-21 13:45:37-finetune.py:240-INFO >> Batch 50 of epoch 4/10, average training loss of previous 2 batches: 0.6274715662002563
176
+ 2024-10-21 13:45:38-finetune.py:240-INFO >> Batch 52 of epoch 4/10, average training loss of previous 2 batches: 0.7559156715869904
177
+ 2024-10-21 13:45:39-finetune.py:240-INFO >> Batch 54 of epoch 4/10, average training loss of previous 2 batches: 0.7475820183753967
178
+ 2024-10-21 13:45:40-finetune.py:240-INFO >> Batch 56 of epoch 4/10, average training loss of previous 2 batches: 0.626720517873764
179
+ 2024-10-21 13:45:41-finetune.py:240-INFO >> Batch 58 of epoch 4/10, average training loss of previous 2 batches: 0.6783787608146667
180
+ 2024-10-21 13:45:41-finetune.py:240-INFO >> Batch 60 of epoch 4/10, average training loss of previous 2 batches: 0.7799589335918427
181
+ 2024-10-21 13:45:42-finetune.py:240-INFO >> Batch 62 of epoch 4/10, average training loss of previous 2 batches: 0.6854929625988007
182
+ 2024-10-21 13:45:43-finetune.py:240-INFO >> Batch 64 of epoch 4/10, average training loss of previous 2 batches: 0.8336123526096344
183
+ 2024-10-21 13:45:44-finetune.py:240-INFO >> Batch 66 of epoch 4/10, average training loss of previous 2 batches: 0.7306537926197052
184
+ 2024-10-21 13:45:45-finetune.py:240-INFO >> Batch 68 of epoch 4/10, average training loss of previous 2 batches: 0.7053666114807129
185
+ 2024-10-21 13:45:46-finetune.py:240-INFO >> Batch 70 of epoch 4/10, average training loss of previous 2 batches: 0.7826265394687653
186
+ 2024-10-21 13:45:47-finetune.py:240-INFO >> Batch 72 of epoch 4/10, average training loss of previous 2 batches: 0.5642250180244446
187
+ 2024-10-21 13:45:48-finetune.py:240-INFO >> Batch 74 of epoch 4/10, average training loss of previous 2 batches: 0.7540428340435028
188
+ 2024-10-21 13:45:48-finetune.py:240-INFO >> Batch 76 of epoch 4/10, average training loss of previous 2 batches: 0.6973932385444641
189
+ 2024-10-21 13:45:49-finetune.py:240-INFO >> Batch 78 of epoch 4/10, average training loss of previous 2 batches: 0.6745010018348694
190
+ 2024-10-21 13:45:50-finetune.py:240-INFO >> Batch 80 of epoch 4/10, average training loss of previous 2 batches: 0.9325462579727173
191
+ 2024-10-21 13:45:51-finetune.py:240-INFO >> Batch 82 of epoch 4/10, average training loss of previous 2 batches: 0.8735631704330444
192
+ 2024-10-21 13:45:52-finetune.py:240-INFO >> Batch 84 of epoch 4/10, average training loss of previous 2 batches: 0.7908239364624023
193
+ 2024-10-21 13:45:53-finetune.py:240-INFO >> Batch 86 of epoch 4/10, average training loss of previous 2 batches: 0.7438755333423615
194
+ 2024-10-21 13:45:54-finetune.py:240-INFO >> Batch 88 of epoch 4/10, average training loss of previous 2 batches: 0.830750972032547
195
+ 2024-10-21 13:45:55-finetune.py:240-INFO >> Batch 90 of epoch 4/10, average training loss of previous 2 batches: 0.8026508390903473
196
+ 2024-10-21 13:45:55-finetune.py:240-INFO >> Batch 92 of epoch 4/10, average training loss of previous 2 batches: 0.7958900928497314
197
+ 2024-10-21 13:45:56-finetune.py:240-INFO >> Batch 94 of epoch 4/10, average training loss of previous 2 batches: 0.7622414231300354
198
+ 2024-10-21 13:45:57-finetune.py:240-INFO >> Batch 96 of epoch 4/10, average training loss of previous 2 batches: 0.8120144009590149
199
+ 2024-10-21 13:45:58-finetune.py:240-INFO >> Batch 98 of epoch 4/10, average training loss of previous 2 batches: 0.7778138220310211
200
+ 2024-10-21 13:45:59-finetune.py:240-INFO >> Batch 100 of epoch 4/10, average training loss of previous 2 batches: 0.8295200765132904
201
+ 2024-10-21 13:46:00-finetune.py:240-INFO >> Batch 2 of epoch 5/10, average training loss of previous 2 batches: 0.6343702375888824
202
+ 2024-10-21 13:46:01-finetune.py:240-INFO >> Batch 4 of epoch 5/10, average training loss of previous 2 batches: 0.7222740948200226
203
+ 2024-10-21 13:46:02-finetune.py:240-INFO >> Batch 6 of epoch 5/10, average training loss of previous 2 batches: 0.6714920699596405
204
+ 2024-10-21 13:46:02-finetune.py:240-INFO >> Batch 8 of epoch 5/10, average training loss of previous 2 batches: 0.9150407016277313
205
+ 2024-10-21 13:46:03-finetune.py:240-INFO >> Batch 10 of epoch 5/10, average training loss of previous 2 batches: 0.6372408270835876
206
+ 2024-10-21 13:46:04-finetune.py:240-INFO >> Batch 12 of epoch 5/10, average training loss of previous 2 batches: 0.7676155269145966
207
+ 2024-10-21 13:46:05-finetune.py:240-INFO >> Batch 14 of epoch 5/10, average training loss of previous 2 batches: 0.8106114268302917
208
+ 2024-10-21 13:46:06-finetune.py:240-INFO >> Batch 16 of epoch 5/10, average training loss of previous 2 batches: 0.8868626058101654
209
+ 2024-10-21 13:46:07-finetune.py:240-INFO >> Batch 18 of epoch 5/10, average training loss of previous 2 batches: 0.766834557056427
210
+ 2024-10-21 13:46:08-finetune.py:240-INFO >> Batch 20 of epoch 5/10, average training loss of previous 2 batches: 0.716394305229187
211
+ 2024-10-21 13:46:09-finetune.py:240-INFO >> Batch 22 of epoch 5/10, average training loss of previous 2 batches: 0.8100038468837738
212
+ 2024-10-21 13:46:09-finetune.py:240-INFO >> Batch 24 of epoch 5/10, average training loss of previous 2 batches: 0.770490825176239
213
+ 2024-10-21 13:46:10-finetune.py:240-INFO >> Batch 26 of epoch 5/10, average training loss of previous 2 batches: 0.7918453514575958
214
+ 2024-10-21 13:46:11-finetune.py:240-INFO >> Batch 28 of epoch 5/10, average training loss of previous 2 batches: 0.6767605841159821
215
+ 2024-10-21 13:46:12-finetune.py:240-INFO >> Batch 30 of epoch 5/10, average training loss of previous 2 batches: 0.6582407057285309
216
+ 2024-10-21 13:46:13-finetune.py:240-INFO >> Batch 32 of epoch 5/10, average training loss of previous 2 batches: 0.6945566833019257
217
+ 2024-10-21 13:46:14-finetune.py:240-INFO >> Batch 34 of epoch 5/10, average training loss of previous 2 batches: 0.7566858530044556
218
+ 2024-10-21 13:46:15-finetune.py:240-INFO >> Batch 36 of epoch 5/10, average training loss of previous 2 batches: 0.8148128092288971
219
+ 2024-10-21 13:46:16-finetune.py:240-INFO >> Batch 38 of epoch 5/10, average training loss of previous 2 batches: 0.6682511568069458
220
+ 2024-10-21 13:46:16-finetune.py:240-INFO >> Batch 40 of epoch 5/10, average training loss of previous 2 batches: 0.6407213807106018
221
+ 2024-10-21 13:46:17-finetune.py:240-INFO >> Batch 42 of epoch 5/10, average training loss of previous 2 batches: 0.6640065908432007
222
+ 2024-10-21 13:46:18-finetune.py:240-INFO >> Batch 44 of epoch 5/10, average training loss of previous 2 batches: 0.7437308430671692
223
+ 2024-10-21 13:46:19-finetune.py:240-INFO >> Batch 46 of epoch 5/10, average training loss of previous 2 batches: 0.7747414410114288
224
+ 2024-10-21 13:46:20-finetune.py:240-INFO >> Batch 48 of epoch 5/10, average training loss of previous 2 batches: 0.7784460186958313
225
+ 2024-10-21 13:46:21-finetune.py:240-INFO >> Batch 50 of epoch 5/10, average training loss of previous 2 batches: 0.5884788930416107
226
+ 2024-10-21 13:46:22-finetune.py:240-INFO >> Batch 52 of epoch 5/10, average training loss of previous 2 batches: 0.6876950860023499
227
+ 2024-10-21 13:46:23-finetune.py:240-INFO >> Batch 54 of epoch 5/10, average training loss of previous 2 batches: 0.6799062788486481
228
+ 2024-10-21 13:46:23-finetune.py:240-INFO >> Batch 56 of epoch 5/10, average training loss of previous 2 batches: 0.5763688087463379
229
+ 2024-10-21 13:46:24-finetune.py:240-INFO >> Batch 58 of epoch 5/10, average training loss of previous 2 batches: 0.6245492696762085
230
+ 2024-10-21 13:46:25-finetune.py:240-INFO >> Batch 60 of epoch 5/10, average training loss of previous 2 batches: 0.7115638852119446
231
+ 2024-10-21 13:46:26-finetune.py:240-INFO >> Batch 62 of epoch 5/10, average training loss of previous 2 batches: 0.6263779103755951
232
+ 2024-10-21 13:46:27-finetune.py:240-INFO >> Batch 64 of epoch 5/10, average training loss of previous 2 batches: 0.7711111307144165
233
+ 2024-10-21 13:46:28-finetune.py:240-INFO >> Batch 66 of epoch 5/10, average training loss of previous 2 batches: 0.6676397919654846
234
+ 2024-10-21 13:46:29-finetune.py:240-INFO >> Batch 68 of epoch 5/10, average training loss of previous 2 batches: 0.6510529518127441
235
+ 2024-10-21 13:46:30-finetune.py:240-INFO >> Batch 70 of epoch 5/10, average training loss of previous 2 batches: 0.719774454832077
236
+ 2024-10-21 13:46:30-finetune.py:240-INFO >> Batch 72 of epoch 5/10, average training loss of previous 2 batches: 0.5127301961183548
237
+ 2024-10-21 13:46:31-finetune.py:240-INFO >> Batch 74 of epoch 5/10, average training loss of previous 2 batches: 0.6937015056610107
238
+ 2024-10-21 13:46:32-finetune.py:240-INFO >> Batch 76 of epoch 5/10, average training loss of previous 2 batches: 0.6372248828411102
239
+ 2024-10-21 13:46:33-finetune.py:240-INFO >> Batch 78 of epoch 5/10, average training loss of previous 2 batches: 0.6238736808300018
240
+ 2024-10-21 13:46:34-finetune.py:240-INFO >> Batch 80 of epoch 5/10, average training loss of previous 2 batches: 0.8558001518249512
241
+ 2024-10-21 13:46:35-finetune.py:240-INFO >> Batch 82 of epoch 5/10, average training loss of previous 2 batches: 0.8063792586326599
242
+ 2024-10-21 13:46:36-finetune.py:240-INFO >> Batch 84 of epoch 5/10, average training loss of previous 2 batches: 0.734971433877945
243
+ 2024-10-21 13:46:37-finetune.py:240-INFO >> Batch 86 of epoch 5/10, average training loss of previous 2 batches: 0.6761626601219177
244
+ 2024-10-21 13:46:37-finetune.py:240-INFO >> Batch 88 of epoch 5/10, average training loss of previous 2 batches: 0.7716780602931976
245
+ 2024-10-21 13:46:38-finetune.py:240-INFO >> Batch 90 of epoch 5/10, average training loss of previous 2 batches: 0.7282805144786835
246
+ 2024-10-21 13:46:39-finetune.py:240-INFO >> Batch 92 of epoch 5/10, average training loss of previous 2 batches: 0.7160176336765289
247
+ 2024-10-21 13:46:40-finetune.py:240-INFO >> Batch 94 of epoch 5/10, average training loss of previous 2 batches: 0.7145473062992096
248
+ 2024-10-21 13:46:41-finetune.py:240-INFO >> Batch 96 of epoch 5/10, average training loss of previous 2 batches: 0.7574485838413239
249
+ 2024-10-21 13:46:42-finetune.py:240-INFO >> Batch 98 of epoch 5/10, average training loss of previous 2 batches: 0.7201968431472778
250
+ 2024-10-21 13:46:43-finetune.py:240-INFO >> Batch 100 of epoch 5/10, average training loss of previous 2 batches: 0.7653315961360931
251
+ 2024-10-21 13:46:44-finetune.py:240-INFO >> Batch 2 of epoch 6/10, average training loss of previous 2 batches: 0.5835276246070862
252
+ 2024-10-21 13:46:44-finetune.py:240-INFO >> Batch 4 of epoch 6/10, average training loss of previous 2 batches: 0.6678467094898224
253
+ 2024-10-21 13:46:45-finetune.py:240-INFO >> Batch 6 of epoch 6/10, average training loss of previous 2 batches: 0.6220656931400299
254
+ 2024-10-21 13:46:46-finetune.py:240-INFO >> Batch 8 of epoch 6/10, average training loss of previous 2 batches: 0.8442187011241913
255
+ 2024-10-21 13:46:47-finetune.py:240-INFO >> Batch 10 of epoch 6/10, average training loss of previous 2 batches: 0.5918219089508057
256
+ 2024-10-21 13:46:48-finetune.py:240-INFO >> Batch 12 of epoch 6/10, average training loss of previous 2 batches: 0.7055420279502869
257
+ 2024-10-21 13:46:49-finetune.py:240-INFO >> Batch 14 of epoch 6/10, average training loss of previous 2 batches: 0.7520930171012878
258
+ 2024-10-21 13:46:50-finetune.py:240-INFO >> Batch 16 of epoch 6/10, average training loss of previous 2 batches: 0.8211362659931183
259
+ 2024-10-21 13:46:51-finetune.py:240-INFO >> Batch 18 of epoch 6/10, average training loss of previous 2 batches: 0.7103014290332794
260
+ 2024-10-21 13:46:51-finetune.py:240-INFO >> Batch 20 of epoch 6/10, average training loss of previous 2 batches: 0.6712160110473633
261
+ 2024-10-21 13:46:52-finetune.py:240-INFO >> Batch 22 of epoch 6/10, average training loss of previous 2 batches: 0.7529181838035583
262
+ 2024-10-21 13:46:53-finetune.py:240-INFO >> Batch 24 of epoch 6/10, average training loss of previous 2 batches: 0.7083960473537445
263
+ 2024-10-21 13:46:54-finetune.py:240-INFO >> Batch 26 of epoch 6/10, average training loss of previous 2 batches: 0.7322015464305878
264
+ 2024-10-21 13:46:55-finetune.py:240-INFO >> Batch 28 of epoch 6/10, average training loss of previous 2 batches: 0.6274860203266144
265
+ 2024-10-21 13:46:56-finetune.py:240-INFO >> Batch 30 of epoch 6/10, average training loss of previous 2 batches: 0.6060699224472046
266
+ 2024-10-21 13:46:57-finetune.py:240-INFO >> Batch 32 of epoch 6/10, average training loss of previous 2 batches: 0.6468688249588013
267
+ 2024-10-21 13:46:58-finetune.py:240-INFO >> Batch 34 of epoch 6/10, average training loss of previous 2 batches: 0.7065079808235168
268
+ 2024-10-21 13:46:58-finetune.py:240-INFO >> Batch 36 of epoch 6/10, average training loss of previous 2 batches: 0.7514010667800903
269
+ 2024-10-21 13:46:59-finetune.py:240-INFO >> Batch 38 of epoch 6/10, average training loss of previous 2 batches: 0.6199988424777985
270
+ 2024-10-21 13:47:00-finetune.py:240-INFO >> Batch 40 of epoch 6/10, average training loss of previous 2 batches: 0.5867652893066406
271
+ 2024-10-21 13:47:01-finetune.py:240-INFO >> Batch 42 of epoch 6/10, average training loss of previous 2 batches: 0.6100884974002838
272
+ 2024-10-21 13:47:02-finetune.py:240-INFO >> Batch 44 of epoch 6/10, average training loss of previous 2 batches: 0.688291072845459
273
+ 2024-10-21 13:47:03-finetune.py:240-INFO >> Batch 46 of epoch 6/10, average training loss of previous 2 batches: 0.7133190035820007
274
+ 2024-10-21 13:47:04-finetune.py:240-INFO >> Batch 48 of epoch 6/10, average training loss of previous 2 batches: 0.7070374190807343
275
+ 2024-10-21 13:47:05-finetune.py:240-INFO >> Batch 50 of epoch 6/10, average training loss of previous 2 batches: 0.5412555932998657
276
+ 2024-10-21 13:47:05-finetune.py:240-INFO >> Batch 52 of epoch 6/10, average training loss of previous 2 batches: 0.6222940981388092
277
+ 2024-10-21 13:47:06-finetune.py:240-INFO >> Batch 54 of epoch 6/10, average training loss of previous 2 batches: 0.6158738136291504
278
+ 2024-10-21 13:47:07-finetune.py:240-INFO >> Batch 56 of epoch 6/10, average training loss of previous 2 batches: 0.5234492123126984
279
+ 2024-10-21 13:47:08-finetune.py:240-INFO >> Batch 58 of epoch 6/10, average training loss of previous 2 batches: 0.5703796744346619
280
+ 2024-10-21 13:47:09-finetune.py:240-INFO >> Batch 60 of epoch 6/10, average training loss of previous 2 batches: 0.6404681205749512
281
+ 2024-10-21 13:47:10-finetune.py:240-INFO >> Batch 62 of epoch 6/10, average training loss of previous 2 batches: 0.5724955499172211
282
+ 2024-10-21 13:47:11-finetune.py:240-INFO >> Batch 64 of epoch 6/10, average training loss of previous 2 batches: 0.7094117403030396
283
+ 2024-10-21 13:47:12-finetune.py:240-INFO >> Batch 66 of epoch 6/10, average training loss of previous 2 batches: 0.6026865541934967
284
+ 2024-10-21 13:47:12-finetune.py:240-INFO >> Batch 68 of epoch 6/10, average training loss of previous 2 batches: 0.5966612547636032
285
+ 2024-10-21 13:47:13-finetune.py:240-INFO >> Batch 70 of epoch 6/10, average training loss of previous 2 batches: 0.654717355966568
286
+ 2024-10-21 13:47:14-finetune.py:240-INFO >> Batch 72 of epoch 6/10, average training loss of previous 2 batches: 0.46248213946819305
287
+ 2024-10-21 13:47:15-finetune.py:240-INFO >> Batch 74 of epoch 6/10, average training loss of previous 2 batches: 0.6266474723815918
288
+ 2024-10-21 13:47:16-finetune.py:240-INFO >> Batch 76 of epoch 6/10, average training loss of previous 2 batches: 0.5776263475418091
289
+ 2024-10-21 13:47:17-finetune.py:240-INFO >> Batch 78 of epoch 6/10, average training loss of previous 2 batches: 0.5670168101787567
290
+ 2024-10-21 13:47:18-finetune.py:240-INFO >> Batch 80 of epoch 6/10, average training loss of previous 2 batches: 0.7725215256214142
291
+ 2024-10-21 13:47:19-finetune.py:240-INFO >> Batch 82 of epoch 6/10, average training loss of previous 2 batches: 0.737803190946579
292
+ 2024-10-21 13:47:19-finetune.py:240-INFO >> Batch 84 of epoch 6/10, average training loss of previous 2 batches: 0.6771756410598755
293
+ 2024-10-21 13:47:20-finetune.py:240-INFO >> Batch 86 of epoch 6/10, average training loss of previous 2 batches: 0.6052446365356445
294
+ 2024-10-21 13:47:21-finetune.py:240-INFO >> Batch 88 of epoch 6/10, average training loss of previous 2 batches: 0.7067574262619019
295
+ 2024-10-21 13:47:22-finetune.py:240-INFO >> Batch 90 of epoch 6/10, average training loss of previous 2 batches: 0.6441330015659332
296
+ 2024-10-21 13:47:23-finetune.py:240-INFO >> Batch 92 of epoch 6/10, average training loss of previous 2 batches: 0.6359881162643433
297
+ 2024-10-21 13:47:24-finetune.py:240-INFO >> Batch 94 of epoch 6/10, average training loss of previous 2 batches: 0.6587474346160889
298
+ 2024-10-21 13:47:25-finetune.py:240-INFO >> Batch 96 of epoch 6/10, average training loss of previous 2 batches: 0.6953521072864532
299
+ 2024-10-21 13:47:26-finetune.py:240-INFO >> Batch 98 of epoch 6/10, average training loss of previous 2 batches: 0.6608266532421112
300
+ 2024-10-21 13:47:26-finetune.py:240-INFO >> Batch 100 of epoch 6/10, average training loss of previous 2 batches: 0.7046919167041779
301
+ 2024-10-21 13:47:27-finetune.py:240-INFO >> Batch 2 of epoch 7/10, average training loss of previous 2 batches: 0.5358765721321106
302
+ 2024-10-21 13:47:28-finetune.py:240-INFO >> Batch 4 of epoch 7/10, average training loss of previous 2 batches: 0.6107850074768066
303
+ 2024-10-21 13:47:29-finetune.py:240-INFO >> Batch 6 of epoch 7/10, average training loss of previous 2 batches: 0.5688601732254028
304
+ 2024-10-21 13:47:30-finetune.py:240-INFO >> Batch 8 of epoch 7/10, average training loss of previous 2 batches: 0.7673291265964508
305
+ 2024-10-21 13:47:31-finetune.py:240-INFO >> Batch 10 of epoch 7/10, average training loss of previous 2 batches: 0.5407192707061768
306
+ 2024-10-21 13:47:32-finetune.py:240-INFO >> Batch 12 of epoch 7/10, average training loss of previous 2 batches: 0.6451371908187866
307
+ 2024-10-21 13:47:33-finetune.py:240-INFO >> Batch 14 of epoch 7/10, average training loss of previous 2 batches: 0.6916760802268982
308
+ 2024-10-21 13:47:33-finetune.py:240-INFO >> Batch 16 of epoch 7/10, average training loss of previous 2 batches: 0.7509200572967529
309
+ 2024-10-21 13:47:34-finetune.py:240-INFO >> Batch 18 of epoch 7/10, average training loss of previous 2 batches: 0.6493522524833679
310
+ 2024-10-21 13:47:35-finetune.py:240-INFO >> Batch 20 of epoch 7/10, average training loss of previous 2 batches: 0.6075893342494965
311
+ 2024-10-21 13:47:36-finetune.py:240-INFO >> Batch 22 of epoch 7/10, average training loss of previous 2 batches: 0.6914695501327515
312
+ 2024-10-21 13:47:37-finetune.py:240-INFO >> Batch 24 of epoch 7/10, average training loss of previous 2 batches: 0.6442109644412994
313
+ 2024-10-21 13:47:38-finetune.py:240-INFO >> Batch 26 of epoch 7/10, average training loss of previous 2 batches: 0.6734310984611511
314
+ 2024-10-21 13:47:39-finetune.py:240-INFO >> Batch 28 of epoch 7/10, average training loss of previous 2 batches: 0.5740040689706802
315
+ 2024-10-21 13:47:40-finetune.py:240-INFO >> Batch 30 of epoch 7/10, average training loss of previous 2 batches: 0.5496492981910706
316
+ 2024-10-21 13:47:40-finetune.py:240-INFO >> Batch 32 of epoch 7/10, average training loss of previous 2 batches: 0.5928905308246613
317
+ 2024-10-21 13:47:41-finetune.py:240-INFO >> Batch 34 of epoch 7/10, average training loss of previous 2 batches: 0.650152176618576
318
+ 2024-10-21 13:47:42-finetune.py:240-INFO >> Batch 36 of epoch 7/10, average training loss of previous 2 batches: 0.6886909306049347
319
+ 2024-10-21 13:47:43-finetune.py:240-INFO >> Batch 38 of epoch 7/10, average training loss of previous 2 batches: 0.5623027682304382
320
+ 2024-10-21 13:47:44-finetune.py:240-INFO >> Batch 40 of epoch 7/10, average training loss of previous 2 batches: 0.5302377045154572
321
+ 2024-10-21 13:47:45-finetune.py:240-INFO >> Batch 42 of epoch 7/10, average training loss of previous 2 batches: 0.5541674494743347
322
+ 2024-10-21 13:47:46-finetune.py:240-INFO >> Batch 44 of epoch 7/10, average training loss of previous 2 batches: 0.6216500401496887
323
+ 2024-10-21 13:47:47-finetune.py:240-INFO >> Batch 46 of epoch 7/10, average training loss of previous 2 batches: 0.6431004106998444
324
+ 2024-10-21 13:47:47-finetune.py:240-INFO >> Batch 48 of epoch 7/10, average training loss of previous 2 batches: 0.6332841515541077
325
+ 2024-10-21 13:47:48-finetune.py:240-INFO >> Batch 50 of epoch 7/10, average training loss of previous 2 batches: 0.4922148138284683
326
+ 2024-10-21 13:47:49-finetune.py:240-INFO >> Batch 52 of epoch 7/10, average training loss of previous 2 batches: 0.5581210851669312
327
+ 2024-10-21 13:47:50-finetune.py:240-INFO >> Batch 54 of epoch 7/10, average training loss of previous 2 batches: 0.554036945104599
328
+ 2024-10-21 13:47:51-finetune.py:240-INFO >> Batch 56 of epoch 7/10, average training loss of previous 2 batches: 0.46584491431713104
329
+ 2024-10-21 13:47:52-finetune.py:240-INFO >> Batch 58 of epoch 7/10, average training loss of previous 2 batches: 0.5075838267803192
330
+ 2024-10-21 13:47:53-finetune.py:240-INFO >> Batch 60 of epoch 7/10, average training loss of previous 2 batches: 0.5658158361911774
331
+ 2024-10-21 13:47:54-finetune.py:240-INFO >> Batch 62 of epoch 7/10, average training loss of previous 2 batches: 0.5088649243116379
332
+ 2024-10-21 13:47:54-finetune.py:240-INFO >> Batch 64 of epoch 7/10, average training loss of previous 2 batches: 0.6368188560009003
333
+ 2024-10-21 13:47:55-finetune.py:240-INFO >> Batch 66 of epoch 7/10, average training loss of previous 2 batches: 0.5332675874233246
334
+ 2024-10-21 13:47:56-finetune.py:240-INFO >> Batch 68 of epoch 7/10, average training loss of previous 2 batches: 0.5267028659582138
335
+ 2024-10-21 13:47:57-finetune.py:240-INFO >> Batch 70 of epoch 7/10, average training loss of previous 2 batches: 0.5833128392696381
336
+ 2024-10-21 13:47:58-finetune.py:240-INFO >> Batch 72 of epoch 7/10, average training loss of previous 2 batches: 0.40017683804035187
337
+ 2024-10-21 13:47:59-finetune.py:240-INFO >> Batch 74 of epoch 7/10, average training loss of previous 2 batches: 0.5564847886562347
338
+ 2024-10-21 13:48:00-finetune.py:240-INFO >> Batch 76 of epoch 7/10, average training loss of previous 2 batches: 0.5111153274774551
339
+ 2024-10-21 13:48:01-finetune.py:240-INFO >> Batch 78 of epoch 7/10, average training loss of previous 2 batches: 0.510166272521019
340
+ 2024-10-21 13:48:01-finetune.py:240-INFO >> Batch 80 of epoch 7/10, average training loss of previous 2 batches: 0.679284930229187
341
+ 2024-10-21 13:48:02-finetune.py:240-INFO >> Batch 82 of epoch 7/10, average training loss of previous 2 batches: 0.6608410477638245
342
+ 2024-10-21 13:48:03-finetune.py:240-INFO >> Batch 84 of epoch 7/10, average training loss of previous 2 batches: 0.6204186677932739
343
+ 2024-10-21 13:48:04-finetune.py:240-INFO >> Batch 86 of epoch 7/10, average training loss of previous 2 batches: 0.5293649286031723
344
+ 2024-10-21 13:48:05-finetune.py:240-INFO >> Batch 88 of epoch 7/10, average training loss of previous 2 batches: 0.627184271812439
345
+ 2024-10-21 13:48:06-finetune.py:240-INFO >> Batch 90 of epoch 7/10, average training loss of previous 2 batches: 0.5681014955043793
346
+ 2024-10-21 13:48:07-finetune.py:240-INFO >> Batch 92 of epoch 7/10, average training loss of previous 2 batches: 0.5675705075263977
347
+ 2024-10-21 13:48:08-finetune.py:240-INFO >> Batch 94 of epoch 7/10, average training loss of previous 2 batches: 0.5900659561157227
348
+ 2024-10-21 13:48:08-finetune.py:240-INFO >> Batch 96 of epoch 7/10, average training loss of previous 2 batches: 0.6214489638805389
349
+ 2024-10-21 13:48:09-finetune.py:240-INFO >> Batch 98 of epoch 7/10, average training loss of previous 2 batches: 0.5980262756347656
350
+ 2024-10-21 13:48:10-finetune.py:240-INFO >> Batch 100 of epoch 7/10, average training loss of previous 2 batches: 0.6394449472427368
351
+ 2024-10-21 13:48:11-finetune.py:240-INFO >> Batch 2 of epoch 8/10, average training loss of previous 2 batches: 0.4861665964126587
352
+ 2024-10-21 13:48:12-finetune.py:240-INFO >> Batch 4 of epoch 8/10, average training loss of previous 2 batches: 0.5529729425907135
353
+ 2024-10-21 13:48:13-finetune.py:240-INFO >> Batch 6 of epoch 8/10, average training loss of previous 2 batches: 0.5235056579113007
354
+ 2024-10-21 13:48:14-finetune.py:240-INFO >> Batch 8 of epoch 8/10, average training loss of previous 2 batches: 0.6846783459186554
355
+ 2024-10-21 13:48:15-finetune.py:240-INFO >> Batch 10 of epoch 8/10, average training loss of previous 2 batches: 0.4859624058008194
356
+ 2024-10-21 13:48:15-finetune.py:240-INFO >> Batch 12 of epoch 8/10, average training loss of previous 2 batches: 0.5769771039485931
357
+ 2024-10-21 13:48:16-finetune.py:240-INFO >> Batch 14 of epoch 8/10, average training loss of previous 2 batches: 0.6253271400928497
358
+ 2024-10-21 13:48:17-finetune.py:240-INFO >> Batch 16 of epoch 8/10, average training loss of previous 2 batches: 0.6685461699962616
359
+ 2024-10-21 13:48:18-finetune.py:240-INFO >> Batch 18 of epoch 8/10, average training loss of previous 2 batches: 0.5832309424877167
360
+ 2024-10-21 13:48:19-finetune.py:240-INFO >> Batch 20 of epoch 8/10, average training loss of previous 2 batches: 0.5380215048789978
361
+ 2024-10-21 13:48:20-finetune.py:240-INFO >> Batch 22 of epoch 8/10, average training loss of previous 2 batches: 0.6208047866821289
362
+ 2024-10-21 13:48:21-finetune.py:240-INFO >> Batch 24 of epoch 8/10, average training loss of previous 2 batches: 0.5718256086111069
363
+ 2024-10-21 13:48:22-finetune.py:240-INFO >> Batch 26 of epoch 8/10, average training loss of previous 2 batches: 0.5938026905059814
364
+ 2024-10-21 13:48:22-finetune.py:240-INFO >> Batch 28 of epoch 8/10, average training loss of previous 2 batches: 0.5184406787157059
365
+ 2024-10-21 13:48:23-finetune.py:240-INFO >> Batch 30 of epoch 8/10, average training loss of previous 2 batches: 0.47936880588531494
366
+ 2024-10-21 13:48:24-finetune.py:240-INFO >> Batch 32 of epoch 8/10, average training loss of previous 2 batches: 0.5321025252342224
367
+ 2024-10-21 13:48:25-finetune.py:240-INFO >> Batch 34 of epoch 8/10, average training loss of previous 2 batches: 0.590179055929184
368
+ 2024-10-21 13:48:26-finetune.py:240-INFO >> Batch 36 of epoch 8/10, average training loss of previous 2 batches: 0.6113237738609314
369
+ 2024-10-21 13:48:27-finetune.py:240-INFO >> Batch 38 of epoch 8/10, average training loss of previous 2 batches: 0.4939865618944168
370
+ 2024-10-21 13:48:28-finetune.py:240-INFO >> Batch 40 of epoch 8/10, average training loss of previous 2 batches: 0.4611608684062958
371
+ 2024-10-21 13:48:29-finetune.py:240-INFO >> Batch 42 of epoch 8/10, average training loss of previous 2 batches: 0.48525552451610565
372
+ 2024-10-21 13:48:29-finetune.py:240-INFO >> Batch 44 of epoch 8/10, average training loss of previous 2 batches: 0.5413184463977814
373
+ 2024-10-21 13:48:30-finetune.py:240-INFO >> Batch 46 of epoch 8/10, average training loss of previous 2 batches: 0.5646709501743317
374
+ 2024-10-21 13:48:31-finetune.py:240-INFO >> Batch 48 of epoch 8/10, average training loss of previous 2 batches: 0.5467158555984497
375
+ 2024-10-21 13:48:32-finetune.py:240-INFO >> Batch 50 of epoch 8/10, average training loss of previous 2 batches: 0.43738752603530884
376
+ 2024-10-21 13:48:33-finetune.py:240-INFO >> Batch 52 of epoch 8/10, average training loss of previous 2 batches: 0.48426760733127594
377
+ 2024-10-21 13:48:34-finetune.py:240-INFO >> Batch 54 of epoch 8/10, average training loss of previous 2 batches: 0.48678092658519745
378
+ 2024-10-21 13:48:35-finetune.py:240-INFO >> Batch 56 of epoch 8/10, average training loss of previous 2 batches: 0.39716456830501556
379
+ 2024-10-21 13:48:36-finetune.py:240-INFO >> Batch 58 of epoch 8/10, average training loss of previous 2 batches: 0.43101413547992706
380
+ 2024-10-21 13:48:36-finetune.py:240-INFO >> Batch 60 of epoch 8/10, average training loss of previous 2 batches: 0.48130224645137787
381
+ 2024-10-21 13:48:37-finetune.py:240-INFO >> Batch 62 of epoch 8/10, average training loss of previous 2 batches: 0.43233947455883026
382
+ 2024-10-21 13:48:38-finetune.py:240-INFO >> Batch 64 of epoch 8/10, average training loss of previous 2 batches: 0.551066517829895
383
+ 2024-10-21 13:48:39-finetune.py:240-INFO >> Batch 66 of epoch 8/10, average training loss of previous 2 batches: 0.45367753505706787
384
+ 2024-10-21 13:48:40-finetune.py:240-INFO >> Batch 68 of epoch 8/10, average training loss of previous 2 batches: 0.43592333793640137
385
+ 2024-10-21 13:48:41-finetune.py:240-INFO >> Batch 70 of epoch 8/10, average training loss of previous 2 batches: 0.5107550472021103
386
+ 2024-10-21 13:48:42-finetune.py:240-INFO >> Batch 72 of epoch 8/10, average training loss of previous 2 batches: 0.3210889846086502
387
+ 2024-10-21 13:48:43-finetune.py:240-INFO >> Batch 74 of epoch 8/10, average training loss of previous 2 batches: 0.4529934674501419
388
+ 2024-10-21 13:48:43-finetune.py:240-INFO >> Batch 76 of epoch 8/10, average training loss of previous 2 batches: 0.4303017109632492
389
+ 2024-10-21 13:48:44-finetune.py:240-INFO >> Batch 78 of epoch 8/10, average training loss of previous 2 batches: 0.4270349144935608
390
+ 2024-10-21 13:48:45-finetune.py:240-INFO >> Batch 80 of epoch 8/10, average training loss of previous 2 batches: 0.5657246708869934
391
+ 2024-10-21 13:48:46-finetune.py:240-INFO >> Batch 82 of epoch 8/10, average training loss of previous 2 batches: 0.5790116190910339
392
+ 2024-10-21 13:48:47-finetune.py:240-INFO >> Batch 84 of epoch 8/10, average training loss of previous 2 batches: 0.5355011522769928
393
+ 2024-10-21 13:48:48-finetune.py:240-INFO >> Batch 86 of epoch 8/10, average training loss of previous 2 batches: 0.4346868395805359
394
+ 2024-10-21 13:48:49-finetune.py:240-INFO >> Batch 88 of epoch 8/10, average training loss of previous 2 batches: 0.5409484803676605
395
+ 2024-10-21 13:48:50-finetune.py:240-INFO >> Batch 90 of epoch 8/10, average training loss of previous 2 batches: 0.47924110293388367
396
+ 2024-10-21 13:48:50-finetune.py:240-INFO >> Batch 92 of epoch 8/10, average training loss of previous 2 batches: 0.4809097498655319
397
+ 2024-10-21 13:48:51-finetune.py:240-INFO >> Batch 94 of epoch 8/10, average training loss of previous 2 batches: 0.5057264566421509
398
+ 2024-10-21 13:48:52-finetune.py:240-INFO >> Batch 96 of epoch 8/10, average training loss of previous 2 batches: 0.5394330024719238
399
+ 2024-10-21 13:48:53-finetune.py:240-INFO >> Batch 98 of epoch 8/10, average training loss of previous 2 batches: 0.5305484235286713
400
+ 2024-10-21 13:48:54-finetune.py:240-INFO >> Batch 100 of epoch 8/10, average training loss of previous 2 batches: 0.5695830583572388
401
+ 2024-10-21 13:48:55-finetune.py:240-INFO >> Batch 2 of epoch 9/10, average training loss of previous 2 batches: 0.44237884879112244
402
+ 2024-10-21 13:48:56-finetune.py:240-INFO >> Batch 4 of epoch 9/10, average training loss of previous 2 batches: 0.4790229946374893
403
+ 2024-10-21 13:48:57-finetune.py:240-INFO >> Batch 6 of epoch 9/10, average training loss of previous 2 batches: 0.44741562008857727
404
+ 2024-10-21 13:48:57-finetune.py:240-INFO >> Batch 8 of epoch 9/10, average training loss of previous 2 batches: 0.5887115597724915
405
+ 2024-10-21 13:48:58-finetune.py:240-INFO >> Batch 10 of epoch 9/10, average training loss of previous 2 batches: 0.40649476647377014
406
+ 2024-10-21 13:48:59-finetune.py:240-INFO >> Batch 12 of epoch 9/10, average training loss of previous 2 batches: 0.49968409538269043
407
+ 2024-10-21 13:49:00-finetune.py:240-INFO >> Batch 14 of epoch 9/10, average training loss of previous 2 batches: 0.5478889048099518
408
+ 2024-10-21 13:49:01-finetune.py:240-INFO >> Batch 16 of epoch 9/10, average training loss of previous 2 batches: 0.5751430839300156
409
+ 2024-10-21 13:49:02-finetune.py:240-INFO >> Batch 18 of epoch 9/10, average training loss of previous 2 batches: 0.5027433037757874
410
+ 2024-10-21 13:49:03-finetune.py:240-INFO >> Batch 20 of epoch 9/10, average training loss of previous 2 batches: 0.46165189146995544
411
+ 2024-10-21 13:49:04-finetune.py:240-INFO >> Batch 22 of epoch 9/10, average training loss of previous 2 batches: 0.5430303812026978
412
+ 2024-10-21 13:49:04-finetune.py:240-INFO >> Batch 24 of epoch 9/10, average training loss of previous 2 batches: 0.49410927295684814
413
+ 2024-10-21 13:49:05-finetune.py:240-INFO >> Batch 26 of epoch 9/10, average training loss of previous 2 batches: 0.5107497572898865
414
+ 2024-10-21 13:49:06-finetune.py:240-INFO >> Batch 28 of epoch 9/10, average training loss of previous 2 batches: 0.44823019206523895
415
+ 2024-10-21 13:49:07-finetune.py:240-INFO >> Batch 30 of epoch 9/10, average training loss of previous 2 batches: 0.39417172968387604
416
+ 2024-10-21 13:49:08-finetune.py:240-INFO >> Batch 32 of epoch 9/10, average training loss of previous 2 batches: 0.4584824740886688
417
+ 2024-10-21 13:49:09-finetune.py:240-INFO >> Batch 34 of epoch 9/10, average training loss of previous 2 batches: 0.5107963383197784
418
+ 2024-10-21 13:49:10-finetune.py:240-INFO >> Batch 36 of epoch 9/10, average training loss of previous 2 batches: 0.5222683548927307
419
+ 2024-10-21 13:49:11-finetune.py:240-INFO >> Batch 38 of epoch 9/10, average training loss of previous 2 batches: 0.4061500281095505
420
+ 2024-10-21 13:49:11-finetune.py:240-INFO >> Batch 40 of epoch 9/10, average training loss of previous 2 batches: 0.3827553242444992
421
+ 2024-10-21 13:49:12-finetune.py:240-INFO >> Batch 42 of epoch 9/10, average training loss of previous 2 batches: 0.40545518696308136
422
+ 2024-10-21 13:49:13-finetune.py:240-INFO >> Batch 44 of epoch 9/10, average training loss of previous 2 batches: 0.45334774255752563
423
+ 2024-10-21 13:49:14-finetune.py:240-INFO >> Batch 46 of epoch 9/10, average training loss of previous 2 batches: 0.4850813150405884
424
+ 2024-10-21 13:49:15-finetune.py:240-INFO >> Batch 48 of epoch 9/10, average training loss of previous 2 batches: 0.4533022940158844
425
+ 2024-10-21 13:49:16-finetune.py:240-INFO >> Batch 50 of epoch 9/10, average training loss of previous 2 batches: 0.3548138439655304
426
+ 2024-10-21 13:49:17-finetune.py:240-INFO >> Batch 52 of epoch 9/10, average training loss of previous 2 batches: 0.3971509784460068
427
+ 2024-10-21 13:49:18-finetune.py:240-INFO >> Batch 54 of epoch 9/10, average training loss of previous 2 batches: 0.41640329360961914
428
+ 2024-10-21 13:49:18-finetune.py:240-INFO >> Batch 56 of epoch 9/10, average training loss of previous 2 batches: 0.32541070878505707
429
+ 2024-10-21 13:49:19-finetune.py:240-INFO >> Batch 58 of epoch 9/10, average training loss of previous 2 batches: 0.3746839165687561
430
+ 2024-10-21 13:49:20-finetune.py:240-INFO >> Batch 60 of epoch 9/10, average training loss of previous 2 batches: 0.39958494901657104
431
+ 2024-10-21 13:49:21-finetune.py:240-INFO >> Batch 62 of epoch 9/10, average training loss of previous 2 batches: 0.3565339893102646
432
+ 2024-10-21 13:49:22-finetune.py:240-INFO >> Batch 64 of epoch 9/10, average training loss of previous 2 batches: 0.45016323029994965
433
+ 2024-10-21 13:49:23-finetune.py:240-INFO >> Batch 66 of epoch 9/10, average training loss of previous 2 batches: 0.3586081713438034
434
+ 2024-10-21 13:49:24-finetune.py:240-INFO >> Batch 68 of epoch 9/10, average training loss of previous 2 batches: 0.3396856337785721
435
+ 2024-10-21 13:49:25-finetune.py:240-INFO >> Batch 70 of epoch 9/10, average training loss of previous 2 batches: 0.4208012521266937
436
+ 2024-10-21 13:49:25-finetune.py:240-INFO >> Batch 72 of epoch 9/10, average training loss of previous 2 batches: 0.2421998679637909
437
+ 2024-10-21 13:49:26-finetune.py:240-INFO >> Batch 74 of epoch 9/10, average training loss of previous 2 batches: 0.36714886128902435
438
+ 2024-10-21 13:49:27-finetune.py:240-INFO >> Batch 76 of epoch 9/10, average training loss of previous 2 batches: 0.3384854272007942
439
+ 2024-10-21 13:49:28-finetune.py:240-INFO >> Batch 78 of epoch 9/10, average training loss of previous 2 batches: 0.34868310391902924
440
+ 2024-10-21 13:49:29-finetune.py:240-INFO >> Batch 80 of epoch 9/10, average training loss of previous 2 batches: 0.46110133826732635
441
+ 2024-10-21 13:49:30-finetune.py:240-INFO >> Batch 82 of epoch 9/10, average training loss of previous 2 batches: 0.4703945964574814
442
+ 2024-10-21 13:49:31-finetune.py:240-INFO >> Batch 84 of epoch 9/10, average training loss of previous 2 batches: 0.44025567173957825
443
+ 2024-10-21 13:49:32-finetune.py:240-INFO >> Batch 86 of epoch 9/10, average training loss of previous 2 batches: 0.34290696680545807
444
+ 2024-10-21 13:49:32-finetune.py:240-INFO >> Batch 88 of epoch 9/10, average training loss of previous 2 batches: 0.4461369514465332
445
+ 2024-10-21 13:49:33-finetune.py:240-INFO >> Batch 90 of epoch 9/10, average training loss of previous 2 batches: 0.38705192506313324
446
+ 2024-10-21 13:49:34-finetune.py:240-INFO >> Batch 92 of epoch 9/10, average training loss of previous 2 batches: 0.4243104159832001
447
+ 2024-10-21 13:49:35-finetune.py:240-INFO >> Batch 94 of epoch 9/10, average training loss of previous 2 batches: 0.41809484362602234
448
+ 2024-10-21 13:49:36-finetune.py:240-INFO >> Batch 96 of epoch 9/10, average training loss of previous 2 batches: 0.47193707525730133
449
+ 2024-10-21 13:49:37-finetune.py:240-INFO >> Batch 98 of epoch 9/10, average training loss of previous 2 batches: 0.47036002576351166
450
+ 2024-10-21 13:49:38-finetune.py:240-INFO >> Batch 100 of epoch 9/10, average training loss of previous 2 batches: 0.49740438163280487
451
+ 2024-10-21 13:49:39-finetune.py:240-INFO >> Batch 2 of epoch 10/10, average training loss of previous 2 batches: 0.37895165383815765
452
+ 2024-10-21 13:49:39-finetune.py:240-INFO >> Batch 4 of epoch 10/10, average training loss of previous 2 batches: 0.40976959466934204
453
+ 2024-10-21 13:49:40-finetune.py:240-INFO >> Batch 6 of epoch 10/10, average training loss of previous 2 batches: 0.4165492206811905
454
+ 2024-10-21 13:49:41-finetune.py:240-INFO >> Batch 8 of epoch 10/10, average training loss of previous 2 batches: 0.517926961183548
455
+ 2024-10-21 13:49:42-finetune.py:240-INFO >> Batch 10 of epoch 10/10, average training loss of previous 2 batches: 0.35744038224220276
456
+ 2024-10-21 13:49:43-finetune.py:240-INFO >> Batch 12 of epoch 10/10, average training loss of previous 2 batches: 0.41863881051540375
457
+ 2024-10-21 13:49:44-finetune.py:240-INFO >> Batch 14 of epoch 10/10, average training loss of previous 2 batches: 0.46624137461185455
458
+ 2024-10-21 13:49:45-finetune.py:240-INFO >> Batch 16 of epoch 10/10, average training loss of previous 2 batches: 0.48127393424510956
459
+ 2024-10-21 13:49:46-finetune.py:240-INFO >> Batch 18 of epoch 10/10, average training loss of previous 2 batches: 0.4298028498888016
460
+ 2024-10-21 13:49:46-finetune.py:240-INFO >> Batch 20 of epoch 10/10, average training loss of previous 2 batches: 0.40097755193710327
461
+ 2024-10-21 13:49:47-finetune.py:240-INFO >> Batch 22 of epoch 10/10, average training loss of previous 2 batches: 0.47262363135814667
462
+ 2024-10-21 13:49:48-finetune.py:240-INFO >> Batch 24 of epoch 10/10, average training loss of previous 2 batches: 0.4252665340900421
463
+ 2024-10-21 13:49:49-finetune.py:240-INFO >> Batch 26 of epoch 10/10, average training loss of previous 2 batches: 0.4307839721441269
464
+ 2024-10-21 13:49:50-finetune.py:240-INFO >> Batch 28 of epoch 10/10, average training loss of previous 2 batches: 0.3781176954507828
465
+ 2024-10-21 13:49:51-finetune.py:240-INFO >> Batch 30 of epoch 10/10, average training loss of previous 2 batches: 0.32740722596645355
466
+ 2024-10-21 13:49:52-finetune.py:240-INFO >> Batch 32 of epoch 10/10, average training loss of previous 2 batches: 0.3956973999738693
467
+ 2024-10-21 13:49:53-finetune.py:240-INFO >> Batch 34 of epoch 10/10, average training loss of previous 2 batches: 0.4431634843349457
468
+ 2024-10-21 13:49:53-finetune.py:240-INFO >> Batch 36 of epoch 10/10, average training loss of previous 2 batches: 0.4427264481782913
469
+ 2024-10-21 13:49:54-finetune.py:240-INFO >> Batch 38 of epoch 10/10, average training loss of previous 2 batches: 0.3347105532884598
470
+ 2024-10-21 13:49:55-finetune.py:240-INFO >> Batch 40 of epoch 10/10, average training loss of previous 2 batches: 0.311638668179512
471
+ 2024-10-21 13:49:56-finetune.py:240-INFO >> Batch 42 of epoch 10/10, average training loss of previous 2 batches: 0.3355199694633484
472
+ 2024-10-21 13:49:57-finetune.py:240-INFO >> Batch 44 of epoch 10/10, average training loss of previous 2 batches: 0.35510262846946716
473
+ 2024-10-21 13:49:58-finetune.py:240-INFO >> Batch 46 of epoch 10/10, average training loss of previous 2 batches: 0.39711587131023407
474
+ 2024-10-21 13:49:59-finetune.py:240-INFO >> Batch 48 of epoch 10/10, average training loss of previous 2 batches: 0.3645806759595871
475
+ 2024-10-21 13:50:00-finetune.py:240-INFO >> Batch 50 of epoch 10/10, average training loss of previous 2 batches: 0.28065013885498047
476
+ 2024-10-21 13:50:00-finetune.py:240-INFO >> Batch 52 of epoch 10/10, average training loss of previous 2 batches: 0.33256326615810394
477
+ 2024-10-21 13:50:01-finetune.py:240-INFO >> Batch 54 of epoch 10/10, average training loss of previous 2 batches: 0.3533191382884979
478
+ 2024-10-21 13:50:02-finetune.py:240-INFO >> Batch 56 of epoch 10/10, average training loss of previous 2 batches: 0.2695348188281059
479
+ 2024-10-21 13:50:03-finetune.py:240-INFO >> Batch 58 of epoch 10/10, average training loss of previous 2 batches: 0.2988472878932953
480
+ 2024-10-21 13:50:04-finetune.py:240-INFO >> Batch 60 of epoch 10/10, average training loss of previous 2 batches: 0.3416874259710312
481
+ 2024-10-21 13:50:05-finetune.py:240-INFO >> Batch 62 of epoch 10/10, average training loss of previous 2 batches: 0.30137424170970917
482
+ 2024-10-21 13:50:06-finetune.py:240-INFO >> Batch 64 of epoch 10/10, average training loss of previous 2 batches: 0.35360729694366455
483
+ 2024-10-21 13:50:07-finetune.py:240-INFO >> Batch 66 of epoch 10/10, average training loss of previous 2 batches: 0.287218913435936
484
+ 2024-10-21 13:50:07-finetune.py:240-INFO >> Batch 68 of epoch 10/10, average training loss of previous 2 batches: 0.254774734377861
485
+ 2024-10-21 13:50:08-finetune.py:240-INFO >> Batch 70 of epoch 10/10, average training loss of previous 2 batches: 0.3343530595302582
486
+ 2024-10-21 13:50:09-finetune.py:240-INFO >> Batch 72 of epoch 10/10, average training loss of previous 2 batches: 0.18388479948043823
487
+ 2024-10-21 13:50:10-finetune.py:240-INFO >> Batch 74 of epoch 10/10, average training loss of previous 2 batches: 0.2653023526072502
488
+ 2024-10-21 13:50:11-finetune.py:240-INFO >> Batch 76 of epoch 10/10, average training loss of previous 2 batches: 0.2973957806825638
489
+ 2024-10-21 13:50:12-finetune.py:240-INFO >> Batch 78 of epoch 10/10, average training loss of previous 2 batches: 0.30567221343517303
490
+ 2024-10-21 13:50:13-finetune.py:240-INFO >> Batch 80 of epoch 10/10, average training loss of previous 2 batches: 0.40487077832221985
491
+ 2024-10-21 13:50:14-finetune.py:240-INFO >> Batch 82 of epoch 10/10, average training loss of previous 2 batches: 0.40729472041130066
492
+ 2024-10-21 13:50:14-finetune.py:240-INFO >> Batch 84 of epoch 10/10, average training loss of previous 2 batches: 0.365250363945961
493
+ 2024-10-21 13:50:15-finetune.py:240-INFO >> Batch 86 of epoch 10/10, average training loss of previous 2 batches: 0.28013309836387634
494
+ 2024-10-21 13:50:16-finetune.py:240-INFO >> Batch 88 of epoch 10/10, average training loss of previous 2 batches: 0.3896760493516922
495
+ 2024-10-21 13:50:17-finetune.py:240-INFO >> Batch 90 of epoch 10/10, average training loss of previous 2 batches: 0.3475548326969147
496
+ 2024-10-21 13:50:18-finetune.py:240-INFO >> Batch 92 of epoch 10/10, average training loss of previous 2 batches: 0.35538066923618317
497
+ 2024-10-21 13:50:19-finetune.py:240-INFO >> Batch 94 of epoch 10/10, average training loss of previous 2 batches: 0.37024180591106415
498
+ 2024-10-21 13:50:20-finetune.py:240-INFO >> Batch 96 of epoch 10/10, average training loss of previous 2 batches: 0.3847298175096512
499
+ 2024-10-21 13:50:21-finetune.py:240-INFO >> Batch 98 of epoch 10/10, average training loss of previous 2 batches: 0.4170578420162201
500
+ 2024-10-21 13:50:21-finetune.py:240-INFO >> Batch 100 of epoch 10/10, average training loss of previous 2 batches: 0.44036656618118286
501
+ 2024-10-21 13:50:30-finetune.py:118-INFO >> chat template saved in train_output/20241021134301/chat_template.json
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.48145466,
8
+ 0.4578275,
9
+ 0.40821073
10
+ ],
11
+ "image_processor_type": "Qwen2VLImageProcessor",
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "max_pixels": 401408,
18
+ "merge_size": 2,
19
+ "min_pixels": 200704,
20
+ "patch_size": 14,
21
+ "processor_class": "Qwen2VLProcessor",
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "max_pixels": 12845056,
26
+ "min_pixels": 3136
27
+ },
28
+ "temporal_patch_size": 2
29
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f33787292af226c4a4842be48a0e614d9524e25dc248e48bb1af0593de5564f9
3
+ size 11420539
tokenizer_config.json ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "151646": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "151647": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "151648": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "151649": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "151650": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "151651": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "151652": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "151653": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "151654": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "151655": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "151656": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ }
116
+ },
117
+ "additional_special_tokens": [
118
+ "<|im_start|>",
119
+ "<|im_end|>",
120
+ "<|object_ref_start|>",
121
+ "<|object_ref_end|>",
122
+ "<|box_start|>",
123
+ "<|box_end|>",
124
+ "<|quad_start|>",
125
+ "<|quad_end|>",
126
+ "<|vision_start|>",
127
+ "<|vision_end|>",
128
+ "<|vision_pad|>",
129
+ "<|image_pad|>",
130
+ "<|video_pad|>"
131
+ ],
132
+ "bos_token": null,
133
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
134
+ "clean_up_tokenization_spaces": false,
135
+ "eos_token": "<|im_end|>",
136
+ "errors": "replace",
137
+ "max_pixels": 401408,
138
+ "min_pixels": 200704,
139
+ "model_max_length": 32768,
140
+ "pad_token": "<|endoftext|>",
141
+ "padding_side": "right",
142
+ "processor_class": "Qwen2VLProcessor",
143
+ "split_special_tokens": false,
144
+ "tokenizer_class": "Qwen2Tokenizer",
145
+ "unk_token": null
146
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff