Hugging Face's logo

zhuokai
/
gpg_baseline_temperature_1.0_Qwen2.5-Math-1.5B_zzk

Model card Files
xet
Community
zhuokai commited on
Commit
74f6b88
·
verified ·
1 Parent(s): e4bb311

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +5 -0
  2. global_step_10/data.pt +3 -0
  3. global_step_100/data.pt +3 -0
  4. global_step_110/data.pt +3 -0
  5. global_step_120/data.pt +3 -0
  6. global_step_130/data.pt +3 -0
  7. global_step_140/data.pt +3 -0
  8. global_step_150/data.pt +3 -0
  9. global_step_160/data.pt +3 -0
  10. global_step_170/data.pt +3 -0
  11. global_step_180/data.pt +3 -0
  12. global_step_190/data.pt +3 -0
  13. global_step_20/data.pt +3 -0
  14. global_step_200/data.pt +3 -0
  15. global_step_210/data.pt +3 -0
  16. global_step_220/data.pt +3 -0
  17. global_step_230/data.pt +3 -0
  18. global_step_240/data.pt +3 -0
  19. global_step_250/data.pt +3 -0
  20. global_step_260/data.pt +3 -0
  21. global_step_270/data.pt +3 -0
  22. global_step_280/data.pt +3 -0
  23. global_step_290/data.pt +3 -0
  24. global_step_30/data.pt +3 -0
  25. global_step_300/data.pt +3 -0
  26. global_step_310/data.pt +3 -0
  27. global_step_320/data.pt +3 -0
  28. global_step_330/data.pt +3 -0
  29. global_step_340/data.pt +3 -0
  30. global_step_350/data.pt +3 -0
  31. global_step_360/data.pt +3 -0
  32. global_step_370/data.pt +3 -0
  33. global_step_380/data.pt +3 -0
  34. global_step_390/data.pt +3 -0
  35. global_step_40/data.pt +3 -0
  36. global_step_400/data.pt +3 -0
  37. global_step_410/actor/extra_state_world_size_8_rank_0.pt +3 -0
  38. global_step_410/actor/extra_state_world_size_8_rank_1.pt +3 -0
  39. global_step_410/actor/extra_state_world_size_8_rank_2.pt +3 -0
  40. global_step_410/actor/extra_state_world_size_8_rank_3.pt +3 -0
  41. global_step_410/actor/extra_state_world_size_8_rank_4.pt +3 -0
  42. global_step_410/actor/extra_state_world_size_8_rank_5.pt +3 -0
  43. global_step_410/actor/extra_state_world_size_8_rank_6.pt +3 -0
  44. global_step_410/actor/extra_state_world_size_8_rank_7.pt +3 -0
  45. global_step_410/actor/fsdp_config.json +4 -0
  46. global_step_410/actor/huggingface/added_tokens.json +24 -0
  47. global_step_410/actor/huggingface/config.json +30 -0
  48. global_step_410/actor/huggingface/generation_config.json +6 -0
  49. global_step_410/actor/huggingface/merges.txt +0 -0
  50. global_step_410/actor/huggingface/special_tokens_map.json +31 -0
.gitattributes CHANGED
@@ -33,3 +33,8 @@ 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
+ global_step_410/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ global_step_420/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ global_step_430/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ global_step_440/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
40
+ global_step_444/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
global_step_10/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87b0aa68bb90a52ccac2e42fe2fa4b7782d0dc60fe7c2106f4814448cae0c49f
3
+ size 1492
global_step_100/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23d364487989cab2153a4854b7c12091cf09a51d7adc278b871fc20e67a0c1a9
3
+ size 1492
global_step_110/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7f18f274ea28d4f24aeaa4289af00529f4d7411898de9aba9afa886f9dfcf94
3
+ size 1492
global_step_120/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76146ebe0c776ec111d002ed718fd2f2bad5993339e7921f4e8393907efc6cd8
3
+ size 1492
global_step_130/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f9a29025a9b84fd388ed49132d2b4dc2d440a59a2e39d544d4147a07d2917f3
3
+ size 1492
global_step_140/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3eeede92110ce3cc91f8dfdc54c380a1ca620a9b913152b4a10e19448ab991d1
3
+ size 1492
global_step_150/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5efda85e305d9f6c4fd22011a05564d07a1f4d21fc54886f936279307ad45099
3
+ size 1492
global_step_160/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3396e20ab2abdea9b84650ae6ed2c3ffd3eeea878eab1b8179c4c7f24b445ccd
3
+ size 1492
global_step_170/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba8fab902b17cde6a574eccb937546f93cf3867fea5ff6768ad37191e653f7d4
3
+ size 1492
global_step_180/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afcf91c8341f8d20a02ffd21a2269b35a2a4bd89916523510c0603810ee123fb
3
+ size 1492
global_step_190/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2552cedda8d6dee503b342bc6aad3b0c028fa70d0b445ff9fdbd3fa4cf1a53b0
3
+ size 1492
global_step_20/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50b31a45f5a2ef59125847b110f05a6c3f801682858905a7d6f3e502dfcb8bfd
3
+ size 1492
global_step_200/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4860ab5dd6f57c0caa1a9e317fae6e5e3ea9ff974c972124c1d178b572cba0b1
3
+ size 1492
global_step_210/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64d0fb77586f833c83e9b3916d557e05d96cb0c1a8fca74ec128abb4d213a5fd
3
+ size 1492
global_step_220/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e43698a50f9e8990a4bd21eaf6ab9243f03d8b03ea154d8d84a14b1ab48c6b3f
3
+ size 1492
global_step_230/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6d4538685f066b0d0faf9d3e38b2f823c60387654c259d3abd70c658799457a
3
+ size 1492
global_step_240/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81326ed33b00622aaf7abf82a0719f18f2beb2ccfb2d2482aa6ef22207119b2a
3
+ size 1492
global_step_250/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4706882108da95c1899c6ae4fd668af3febe8b74dc9fb101a2b3b7ca43ab1928
3
+ size 1492
global_step_260/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bb7110be88cb7448b150624554ee5032874c18f0ffdc9c5e3084b58b8d561a2
3
+ size 1556
global_step_270/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48d718698e4645cc052ff4039d38d9a94bc81da2d74e86ed5082f47653ea5026
3
+ size 1556
global_step_280/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b6e03acd45bc63c183a160108b56d4821a952932ac8d4a2ac93a9642b58027b
3
+ size 1556
global_step_290/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:638681405ac70a5536f8e399ecec5fcfb3542b7239413997cbd918e2a81d3da8
3
+ size 1556
global_step_30/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:443fe897b0bb45c835d15686fade59b2efbd548b60c8294610f01bb0a9b3952e
3
+ size 1492
global_step_300/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:471573bcfdecfbf01725146f385f3564c9e6defcca299c7388ffe0d4620bec8d
3
+ size 1556
global_step_310/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cd585a5341af8faf745d5f5266ccfc70ce09d56d8c4944cb033d86358693679
3
+ size 1556
global_step_320/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:536ea27d99d60281c5389a0d90e0e33c92aba0e68d31997e857ea59f805b4448
3
+ size 1556
global_step_330/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b8fb6d7c834d51b5ddbd8883b2b99b2f164295ae4ce7e89c6b999cc30e82a13
3
+ size 1556
global_step_340/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27d6c35fd43de913d24b2ed711c280f8e7430f0af469b94540023df18f9215f6
3
+ size 1556
global_step_350/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4f97cf3d7573d784154b4eb8221725ccc428c7013aba046e57f3539e674f040
3
+ size 1556
global_step_360/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e388f9cb72c38d1f20b61bcc4a21513ea253942f445d8d5cd5b7e0dd443726be
3
+ size 1556
global_step_370/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83515848258f40b79a55be4edde240c6e11d02ebfce2f91a39c89d317aa4c7da
3
+ size 1556
global_step_380/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6a9e32f976a7b2bfdab5b025fd789c279815c1b61b36828a8e27547bd38db0b
3
+ size 1556
global_step_390/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:695b285bc5c5f9ae786364a349f507f7201a0c729a3c455a548ff621016463d3
3
+ size 1556
global_step_40/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58e66772ba45823ecbe29b65ae135c49d126b3db8f36e42f3755b1593c68cd72
3
+ size 1492
global_step_400/data.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3e912830f90c7a9f345d4e8e1963d0ca1964e6e73d14af2c03c452e22ce7b6c
3
+ size 1556
global_step_410/actor/extra_state_world_size_8_rank_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05b3ba55c9ab6f3006692a7c6183587902010dfd80b4ec0dde50c9a56f489b20
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_1.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:936d9887d3ff3688cc3a2ea9298b8ca0531f83b612463829b9dee248647925d6
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:408538fda8acacabcdec600c3b52d58f33fc116146fdfff08ad8dad398f2c7dc
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_3.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2da863bf61902e45b320b63a58adebeda67a457be79b67057d0e53591ac9f89
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cecd819e59aee6a4a807036b2067ac751192cf69b917a5a598adb6c2bc67651b
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_5.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97b4f0ec301ea549313d6f189a6078cc8e76af2155a10640ff7349f6710521f7
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_6.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:637234e33bd402a548e946634ebe04fd7f62954964ee7fe36933221d2fdab47c
3
+ size 14632
global_step_410/actor/extra_state_world_size_8_rank_7.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b18f4e1215deeeab897b1128842ca688e1bc80566906e3b5834f2445d60675f4
3
+ size 14632
global_step_410/actor/fsdp_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "FSDP_version": 1,
3
+ "world_size": 8
4
+ }
global_step_410/actor/huggingface/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
global_step_410/actor/huggingface/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "eos_token_id": 151643,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 1536,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 8960,
12
+ "max_position_embeddings": 4096,
13
+ "max_window_layers": 21,
14
+ "model_type": "qwen2",
15
+ "num_attention_heads": 12,
16
+ "num_hidden_layers": 28,
17
+ "num_key_value_heads": 2,
18
+ "pad_token_id": 151643,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_scaling": null,
21
+ "rope_theta": 10000,
22
+ "sliding_window": 4096,
23
+ "tie_word_embeddings": true,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.51.1",
26
+ "use_cache": true,
27
+ "use_mrope": false,
28
+ "use_sliding_window": false,
29
+ "vocab_size": 151936
30
+ }
global_step_410/actor/huggingface/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_new_tokens": 2048,
5
+ "transformers_version": "4.51.1"
6
+ }
global_step_410/actor/huggingface/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
global_step_410/actor/huggingface/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": "<|endoftext|>",
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
+ }