Hugging Face's logo

mrtuandao
/
dunono

Safetensors
Model card Files
xet
Community
mrtuandao commited on
Commit
7c4dc81
·
verified ·
1 Parent(s): aa4a918

Upload folder using huggingface_hub

Browse files
Files changed (23) hide show
  1. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/config.json +39 -0
  2. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/generation_config.json +6 -0
  3. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/lr_scheduler.pt +3 -0
  4. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/merges.txt +0 -0
  5. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/model.safetensors +3 -0
  6. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/optimizer.pt +3 -0
  7. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_0.pt +3 -0
  8. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_1.pt +3 -0
  9. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_2.pt +3 -0
  10. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_3.pt +3 -0
  11. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_4.pt +3 -0
  12. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_5.pt +3 -0
  13. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_6.pt +3 -0
  14. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_7.pt +3 -0
  15. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/query_projector.pt +3 -0
  16. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/s2t_projector.pt +3 -0
  17. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/special_tokens_map.json +6 -0
  18. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/t2s_projector.pt +3 -0
  19. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/tokenizer.json +0 -0
  20. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/tokenizer_config.json +21 -0
  21. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/vocab.json +0 -0
  22. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/tuandao_qwen1.5-1.8b_to_gpt2-120m.log +52 -0
  23. experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/tuandao_qwen1.5-1.8b_to_gpt2-120m_metrics.jsonl +0 -0
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "dtype": "float32",
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 50256,
11
+ "initializer_range": 0.02,
12
+ "layer_norm_epsilon": 1e-05,
13
+ "model_type": "gpt2",
14
+ "n_ctx": 1024,
15
+ "n_embd": 768,
16
+ "n_head": 12,
17
+ "n_inner": null,
18
+ "n_layer": 12,
19
+ "n_positions": 1024,
20
+ "pad_token_id": 50256,
21
+ "reorder_and_upcast_attn": false,
22
+ "resid_pdrop": 0.1,
23
+ "scale_attn_by_inverse_layer_idx": false,
24
+ "scale_attn_weights": true,
25
+ "summary_activation": null,
26
+ "summary_first_dropout": 0.1,
27
+ "summary_proj_to_labels": true,
28
+ "summary_type": "cls_index",
29
+ "summary_use_proj": true,
30
+ "task_specific_params": {
31
+ "text-generation": {
32
+ "do_sample": true,
33
+ "max_length": 50
34
+ }
35
+ },
36
+ "transformers_version": "4.56.0",
37
+ "use_cache": true,
38
+ "vocab_size": 50257
39
+ }
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.56.0"
6
+ }
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/lr_scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cfac693e0b5af466e48332e041e057ee27abaf9cab8eb3a91852f357a818c75
3
+ size 1483
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9be8d6d89ed7b0ed28ec0e4e6bcfcf5a45e352ca1197c30feb3d0722fc2b146c
3
+ size 497774208
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbfa3917cd670378b48ea1dba7de94800f133d7cf77da4f8f1ab161473bca23b
3
+ size 1171926795
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc955644541767ab0d806b111eafa4081220da86e680f16971055f6fa5f8840c
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d63e158e13360f720ef3daf6dd2897fcef7b537cbc397b7b395583a4ae1ca762
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3957878aee0750d99156b29e6563bb9dcaca227d9a7d87b66731ea9ac8cc0c85
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:544603d06e4a9b3fdddddf8e6ec560376809fbfdb248f5b659afe9accf9641a6
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e297cbbbb65e8f7a0d6b079feede5d4edf9fcd1f01579cb3d02ac085ccc9127e
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_5.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84819519aded0aecf8ae78f55e84d1607b866e93dcd4e149083bbc9e8b6d4fb2
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_6.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42377842e17e1447dd50808ef2b53585786e6a417d5a179d763604a3ffd20b00
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/projector_7.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:046c0e63bf3878e006cd23b3bd90b373ff83a0403848ee0e30b273258215436b
3
+ size 6296517
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/query_projector.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42df0c0ae02b8e1d938f1afc4fc610130bb6acc5dee1ec4fded18b63359532d2
3
+ size 25184229
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/s2t_projector.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf34a7e46d3b54ac2932bcce97d254d1748b8658c53dd15e01bb3513f38343c0
3
+ size 6301653
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|endoftext|>",
5
+ "unk_token": "<|endoftext|>"
6
+ }
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/t2s_projector.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59a3a99a2aefb864dc3a4e32f3ee301be4fa65548b774d45acd09f43dac410f7
3
+ size 6296533
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/tokenizer_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "pad_token": "<|endoftext|>",
19
+ "tokenizer_class": "GPT2Tokenizer",
20
+ "unk_token": "<|endoftext|>"
21
+ }
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/checkpoints/epoch_12/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/tuandao_qwen1.5-1.8b_to_gpt2-120m.log CHANGED
@@ -645,3 +645,55 @@
645
  2025-12-09 06:01:21,749 - absl - INFO - Using default tokenizer.
646
  2025-12-09 06:01:24,709 - root - INFO - Epoch 12/20 eval rougeL: 0.21586395602741776
647
  2025-12-09 06:01:26,943 - root - INFO - Epoch 13/20
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
645
  2025-12-09 06:01:21,749 - absl - INFO - Using default tokenizer.
646
  2025-12-09 06:01:24,709 - root - INFO - Epoch 12/20 eval rougeL: 0.21586395602741776
647
  2025-12-09 06:01:26,943 - root - INFO - Epoch 13/20
648
+ 2025-12-09 06:03:23,167 - absl - INFO - Using default tokenizer.
649
+ 2025-12-09 06:03:26,056 - root - INFO - Step 17281/28600 train rougeL: 0.8877341920374706
650
+ 2025-12-09 06:03:26,365 - root - INFO - Step 17281/28600 loss: 1.1440061330795288, nll_loss: 0.12626519799232483, distill_loss: 0.1744195520877838, dskd_loss: 1.8610622882843018
651
+ 2025-12-09 06:05:30,510 - absl - INFO - Using default tokenizer.
652
+ 2025-12-09 06:05:33,754 - root - INFO - Step 17409/28600 train rougeL: 0.7394414648471688
653
+ 2025-12-09 06:05:34,049 - root - INFO - Step 17409/28600 loss: 1.007826328277588, nll_loss: 0.1303185373544693, distill_loss: 0.17187200486660004, dskd_loss: 1.5831435918807983
654
+ 2025-12-09 06:07:28,268 - absl - INFO - Using default tokenizer.
655
+ 2025-12-09 06:07:29,628 - root - INFO - Step 17537/28600 train rougeL: 0.8919871794871795
656
+ 2025-12-09 06:07:29,910 - root - INFO - Step 17537/28600 loss: 0.8760161399841309, nll_loss: 0.09825839102268219, distill_loss: 0.15972398221492767, dskd_loss: 1.3957915306091309
657
+ 2025-12-09 06:09:29,699 - absl - INFO - Using default tokenizer.
658
+ 2025-12-09 06:09:31,518 - root - INFO - Step 17665/28600 train rougeL: 0.875
659
+ 2025-12-09 06:09:31,800 - root - INFO - Step 17665/28600 loss: 0.9362841248512268, nll_loss: 0.09939136356115341, distill_loss: 0.15755969285964966, dskd_loss: 1.516225814819336
660
+ 2025-12-09 06:11:35,868 - absl - INFO - Using default tokenizer.
661
+ 2025-12-09 06:11:38,413 - root - INFO - Step 17793/28600 train rougeL: 0.8717079732028185
662
+ 2025-12-09 06:11:38,716 - root - INFO - Step 17793/28600 loss: 0.9913924932479858, nll_loss: 0.15530525147914886, distill_loss: 0.170277938246727, dskd_loss: 1.5018965005874634
663
+ 2025-12-09 06:13:38,174 - absl - INFO - Using default tokenizer.
664
+ 2025-12-09 06:13:41,233 - root - INFO - Step 17921/28600 train rougeL: 0.8828334311342108
665
+ 2025-12-09 06:13:41,527 - root - INFO - Step 17921/28600 loss: 1.1010695695877075, nll_loss: 0.13439980149269104, distill_loss: 0.2036806046962738, dskd_loss: 1.729658842086792
666
+ 2025-12-09 06:15:41,550 - absl - INFO - Using default tokenizer.
667
+ 2025-12-09 06:15:43,754 - root - INFO - Step 18049/28600 train rougeL: 0.7698166988391756
668
+ 2025-12-09 06:15:44,078 - root - INFO - Step 18049/28600 loss: 1.1418837308883667, nll_loss: 0.16063542664051056, distill_loss: 0.16913819313049316, dskd_loss: 1.7933584451675415
669
+ 2025-12-09 06:17:41,141 - absl - INFO - Using default tokenizer.
670
+ 2025-12-09 06:17:44,373 - root - INFO - Step 18177/28600 train rougeL: 0.8504827126515868
671
+ 2025-12-09 06:17:44,683 - root - INFO - Step 18177/28600 loss: 0.8415998220443726, nll_loss: 0.12076552957296371, distill_loss: 0.1570442020893097, dskd_loss: 1.2846243381500244
672
+ 2025-12-09 06:19:39,203 - absl - INFO - Using default tokenizer.
673
+ 2025-12-09 06:19:41,200 - root - INFO - Step 18305/28600 train rougeL: 0.9
674
+ 2025-12-09 06:19:41,483 - root - INFO - Step 18305/28600 loss: 0.8405715227127075, nll_loss: 0.10600381344556808, distill_loss: 0.15056058764457703, dskd_loss: 1.3185747861862183
675
+ 2025-12-09 06:21:38,991 - absl - INFO - Using default tokenizer.
676
+ 2025-12-09 06:21:41,863 - root - INFO - Step 18433/28600 train rougeL: 0.7330878138255187
677
+ 2025-12-09 06:21:42,176 - root - INFO - Step 18433/28600 loss: 1.2977479696273804, nll_loss: 0.22122560441493988, distill_loss: 0.17732103168964386, dskd_loss: 1.9757237434387207
678
+ 2025-12-09 06:23:39,174 - absl - INFO - Using default tokenizer.
679
+ 2025-12-09 06:23:41,229 - root - INFO - Step 18561/28600 train rougeL: 0.8313110586464664
680
+ 2025-12-09 06:23:41,520 - root - INFO - Step 18561/28600 loss: 1.1198689937591553, nll_loss: 0.19140806794166565, distill_loss: 0.16645745933055878, dskd_loss: 1.6904644966125488
681
+ 2025-12-09 06:24:09,095 - root - INFO - Epoch 13/20 finished
682
+ 2025-12-09 06:24:09,126 - absl - INFO - Using default tokenizer.
683
+ 2025-12-09 06:24:12,406 - absl - INFO - Using default tokenizer.
684
+ 2025-12-09 06:24:15,727 - absl - INFO - Using default tokenizer.
685
+ 2025-12-09 06:24:18,899 - absl - INFO - Using default tokenizer.
686
+ 2025-12-09 06:24:22,027 - absl - INFO - Using default tokenizer.
687
+ 2025-12-09 06:24:25,343 - absl - INFO - Using default tokenizer.
688
+ 2025-12-09 06:24:28,625 - absl - INFO - Using default tokenizer.
689
+ 2025-12-09 06:24:31,908 - absl - INFO - Using default tokenizer.
690
+ 2025-12-09 06:24:35,240 - absl - INFO - Using default tokenizer.
691
+ 2025-12-09 06:24:38,544 - absl - INFO - Using default tokenizer.
692
+ 2025-12-09 06:24:41,807 - absl - INFO - Using default tokenizer.
693
+ 2025-12-09 06:24:45,030 - absl - INFO - Using default tokenizer.
694
+ 2025-12-09 06:24:48,306 - absl - INFO - Using default tokenizer.
695
+ 2025-12-09 06:24:51,557 - absl - INFO - Using default tokenizer.
696
+ 2025-12-09 06:24:54,704 - absl - INFO - Using default tokenizer.
697
+ 2025-12-09 06:24:57,913 - absl - INFO - Using default tokenizer.
698
+ 2025-12-09 06:25:01,092 - root - INFO - Epoch 13/20 eval rougeL: 0.21788657515474516
699
+ 2025-12-09 06:25:03,338 - root - INFO - Epoch 14/20
experiments/tuandao_qwen1.5-1.8b_to_gpt2-120m/20251209_011341/tuandao_qwen1.5-1.8b_to_gpt2-120m_metrics.jsonl CHANGED
The diff for this file is too large to render. See raw diff