Jongbin-kr commited on
Commit
7e641fe
·
verified ·
1 Parent(s): ed8d9b2

Training in progress, epoch 1

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +67 -0
  2. adapter_config.json +48 -0
  3. adapter_model.safetensors +3 -0
  4. chat_template.jinja +4 -0
  5. completions/completions_00001.parquet +3 -0
  6. completions/completions_00002.parquet +3 -0
  7. completions/completions_00003.parquet +3 -0
  8. completions/completions_00004.parquet +3 -0
  9. completions/completions_00005.parquet +3 -0
  10. completions/completions_00006.parquet +3 -0
  11. completions/completions_00007.parquet +3 -0
  12. completions/completions_00008.parquet +3 -0
  13. completions/completions_00009.parquet +3 -0
  14. completions/completions_00010.parquet +3 -0
  15. completions/completions_00011.parquet +3 -0
  16. completions/completions_00012.parquet +3 -0
  17. completions/completions_00013.parquet +3 -0
  18. completions/completions_00014.parquet +3 -0
  19. completions/completions_00015.parquet +3 -0
  20. completions/completions_00016.parquet +3 -0
  21. completions/completions_00017.parquet +3 -0
  22. completions/completions_00018.parquet +3 -0
  23. completions/completions_00019.parquet +3 -0
  24. completions/completions_00020.parquet +3 -0
  25. completions/completions_00021.parquet +3 -0
  26. completions/completions_00022.parquet +3 -0
  27. completions/completions_00023.parquet +3 -0
  28. completions/completions_00024.parquet +3 -0
  29. completions/completions_00025.parquet +3 -0
  30. completions/completions_00026.parquet +3 -0
  31. completions/completions_00027.parquet +3 -0
  32. completions/completions_00028.parquet +3 -0
  33. completions/completions_00029.parquet +3 -0
  34. completions/completions_00030.parquet +3 -0
  35. completions/completions_00031.parquet +3 -0
  36. completions/completions_00032.parquet +3 -0
  37. completions/completions_00033.parquet +3 -0
  38. completions/completions_00034.parquet +3 -0
  39. completions/completions_00035.parquet +3 -0
  40. completions/completions_00036.parquet +3 -0
  41. completions/completions_00037.parquet +3 -0
  42. completions/completions_00038.parquet +3 -0
  43. completions/completions_00039.parquet +3 -0
  44. completions/completions_00040.parquet +3 -0
  45. completions/completions_00041.parquet +3 -0
  46. completions/completions_00042.parquet +3 -0
  47. completions/completions_00043.parquet +3 -0
  48. completions/completions_00044.parquet +3 -0
  49. completions/completions_00045.parquet +3 -0
  50. completions/completions_00046.parquet +3 -0
README.md ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct
3
+ library_name: transformers
4
+ model_name: exaone_7b_verireason_vanilla_grpo-r02e5-hpft-stable
5
+ tags:
6
+ - generated_from_trainer
7
+ - grpo
8
+ - trl
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for exaone_7b_verireason_vanilla_grpo-r02e5-hpft-stable
13
+
14
+ This model is a fine-tuned version of [LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct](https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
23
+ generator = pipeline("text-generation", model="Jongbin-kr/exaone_7b_verireason_vanilla_grpo-r02e5-hpft-stable", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/snu-skiml/verireason-grpo/runs/sglzjif9)
31
+
32
+
33
+
34
+ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
35
+
36
+ ### Framework versions
37
+
38
+ - TRL: 1.6.0
39
+ - Transformers: 5.7.0
40
+ - Pytorch: 2.10.0+cu128
41
+ - Datasets: 5.0.0
42
+ - Tokenizers: 0.22.2
43
+
44
+ ## Citations
45
+
46
+ Cite GRPO as:
47
+
48
+ ```bibtex
49
+ @article{shao2024deepseekmath,
50
+ title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
51
+ author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
52
+ year = 2024,
53
+ eprint = {arXiv:2402.03300},
54
+ }
55
+ ```
56
+
57
+ Cite TRL as:
58
+
59
+ ```bibtex
60
+ @software{vonwerra2020trl,
61
+ title = {{TRL: Transformers Reinforcement Learning}},
62
+ author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
63
+ license = {Apache-2.0},
64
+ url = {https://github.com/huggingface/trl},
65
+ year = {2020}
66
+ }
67
+ ```
adapter_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 128,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "peft_type": "LORA",
27
+ "peft_version": "0.19.1",
28
+ "qalora_group_size": 16,
29
+ "r": 64,
30
+ "rank_pattern": {},
31
+ "revision": null,
32
+ "target_modules": [
33
+ "c_fc_0",
34
+ "q_proj",
35
+ "c_proj",
36
+ "c_fc_1",
37
+ "k_proj",
38
+ "out_proj",
39
+ "v_proj"
40
+ ],
41
+ "target_parameters": null,
42
+ "task_type": "CAUSAL_LM",
43
+ "trainable_token_indices": null,
44
+ "use_bdlora": null,
45
+ "use_dora": false,
46
+ "use_qalora": false,
47
+ "use_rslora": false
48
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b93d74e251d584eee34d36ed262523e2ed9a5a04393f62d8ea6c032397a4f344
3
+ size 671150560
chat_template.jinja ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {% for message in messages %}{% if loop.first and message['role'] != 'system' %}{{ '[|system|][|endofturn|]
2
+ ' }}{% endif %}{{ '[|' + message['role'] + '|]' + message['content'] }}{% if message['role'] == 'user' %}{{ '
3
+ ' }}{% else %}{{ '[|endofturn|]
4
+ ' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '[|assistant|]' }}{% endif %}
completions/completions_00001.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57737fed1c79b8b85d8ec8c5bc47e43fda44674ee46938f864914a8e4950fa37
3
+ size 37352
completions/completions_00002.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:795fe9368732b5c2488e955e39f95b66d19c88c1bb41e201a5730e18f011d71a
3
+ size 40345
completions/completions_00003.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77785aac3f498fb1c852ebb75f164ef9f1a65e9f98e9c3f9b399351be7ed98aa
3
+ size 44393
completions/completions_00004.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4e7815683f65b2c2d8567b1548f1313bd43ba95772c89f121bd90f2a8826369
3
+ size 32481
completions/completions_00005.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f5c6f0a632046962b1e3a7e49b7adf2f071b7549525f36b05910dc394282404
3
+ size 32384
completions/completions_00006.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57443fcfe1de1b7e23abfc3dda6319ef9b7f7132668f0be38b4e8ead0bf5a8e3
3
+ size 34941
completions/completions_00007.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b62aaabf6f5c2a070ba973fb1bb130f13bcabcaa25a091d6e9a2c7db5a918be
3
+ size 40338
completions/completions_00008.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f42f1cf3354449108bb0fb39bb00aed499f1c41d8efc560491b74b0b0e9a3e2c
3
+ size 33797
completions/completions_00009.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:904c61cf99170d5b2117fdab7ce34d4cb105e0fa4a4069dd9c42e665a4a9ebb7
3
+ size 31742
completions/completions_00010.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:435333a5dba1edd16bbd3d5d9caaad483c15a8699f5c064e2666332fcba84edb
3
+ size 33335
completions/completions_00011.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d9e14dc2e1753fe514708703e13c64da36f6ab9d3c50271b2fa2633632fe777
3
+ size 23625
completions/completions_00012.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:183d3e513d276b9221f14c0b2e0e8cc1ccba6e4698240a26d614dcddffe87799
3
+ size 35002
completions/completions_00013.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40d808cd4118f538b49d1e3329974072bb5673dddef6fab3911604a416f6280c
3
+ size 29596
completions/completions_00014.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0642229146288e6108e4d8037d0757aec4916509e57f4b9d5c7ef847fe3d608
3
+ size 30640
completions/completions_00015.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e36b6a3bbe2b1e509ec3c6c562018910d069b3c061932e9cdf80c39536b1c5f
3
+ size 30815
completions/completions_00016.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a97e14ee684f957d2d73042fde23a41a4dd17fcccc8ebaae0f8be755f3f4043
3
+ size 32210
completions/completions_00017.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:794238e47940182c1dc48fea2113ad3cc8a0f13b862bcf9bf9fd59437989a584
3
+ size 32923
completions/completions_00018.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1026762b13314e337f2b98adf41abb8f1d840f77aa153cffce58c4dc6ca90635
3
+ size 34116
completions/completions_00019.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b18cf344bc8f0853673ec114c5a56c132c92ba877b1062f78e3f3480184aa21a
3
+ size 33088
completions/completions_00020.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7e2b2226e02baabb2adacf937604dd9456cf4770dcc20b5cf190a81a48d1607
3
+ size 34056
completions/completions_00021.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c99d96778cad52befc89c49cc14d1657457daf03013ac0277cebb7f49e661a44
3
+ size 31318
completions/completions_00022.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:534be888ef353955418b51080fe659b9d0e82a605ca19c45b6bd4320bd74b829
3
+ size 35867
completions/completions_00023.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fed74dc5c5d84789551d61d51cb489537f31e506462d14a52ee15e23502a37e6
3
+ size 27526
completions/completions_00024.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc2f1633e422f166301fa071e5bdc80fadc5bb050e42eb477e932378c0ecb06a
3
+ size 28745
completions/completions_00025.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ff6cb8de3b11372019953841de2386926b1b54b78787fb55387c146421f6d32
3
+ size 32658
completions/completions_00026.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50f1547295241b6be7b72aaabbd03e5b370e91fd2db8649107126180aa06a5ea
3
+ size 25986
completions/completions_00027.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62a806519437fb5c9407f5c0b7a8496cb7c59d1b91162969ead3af477396b4e9
3
+ size 35839
completions/completions_00028.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ae49c6157b7e211610386bf5178811985ba41e752d35b89d038599d4f2b019e
3
+ size 31477
completions/completions_00029.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c571c3264e1e2b9ebdfaa4526f88c4e2759981c5e099f1cd730b2e93296267d6
3
+ size 36418
completions/completions_00030.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dca449382cd8044a250ebc7079c5c443f22ed7b9c026673c42cd5d45c7b1cb85
3
+ size 32831
completions/completions_00031.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e15d2439c22a74201697ec347a32bb4a917457ad1112cd6bab3a3e7980531ed5
3
+ size 33788
completions/completions_00032.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e81851c3a2b08685d0d233dbfa7a9a8876be197684c0e4fc6a1a67e328c0e48f
3
+ size 40969
completions/completions_00033.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91c3cc72b1fae730c6fb73791d9fb372b97dcae08cfc84ba381616ee455ccbe6
3
+ size 37507
completions/completions_00034.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89191d9746a191af850fe847918bf07fcac37cbc6c011f3ba3ecfd062ef9c2f9
3
+ size 37427
completions/completions_00035.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73ac8c1ede8be018908115e4ae99cdfb425e0eae023afee74d1474c46a3363fe
3
+ size 32572
completions/completions_00036.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f8d58bc12f1ed15fc17505c5f83da3c2dd978324881400d43f0fe191af4ef59
3
+ size 33894
completions/completions_00037.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:148da54c2b583a05b271048b5ee9b6b2b412e642eeba7fdbbbae57b2c21e3ead
3
+ size 27433
completions/completions_00038.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6317eb9a8ebaa3ae125f0e287a617fd42bea8406a06075eeb12239fb0701dc4
3
+ size 32150
completions/completions_00039.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0a6171ecbf627ff5ca7ec99680b165e997da817cb1c2db775d938de6087042e
3
+ size 28860
completions/completions_00040.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5633660cffff57a320b517ab6b8e4fe92ba8b4ab750f3b9384832547772deb94
3
+ size 29892
completions/completions_00041.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd171c3ad0fb78bb3f58c618e44836c8382a13b336451ad6666ea3e4052f3516
3
+ size 31960
completions/completions_00042.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1493a6e6550dd0fa0caca0c6a25b2c372c141e794e0523b37c587408f03daee6
3
+ size 35573
completions/completions_00043.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17f3814527dde32c3f92769fe864565113c46950e4502f2d2e091eb8a029c4fd
3
+ size 28752
completions/completions_00044.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32cf4cad448303e4c9c009acc8ee2214a3d45287c341d0e618af5cd969527059
3
+ size 37036
completions/completions_00045.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:263cd5ac670b03a7d1a8060498dadabbd25277179af1a929f231dda58c50af2b
3
+ size 37774
completions/completions_00046.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9920303969fdece4aa8357e1833dfeec213f2105cf18246bcf0b4690e87ce2a
3
+ size 37627