bensondccnqwc commited on
Commit
d568d68
·
verified ·
1 Parent(s): a99ebfc

Add files using upload-large-folder tool

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 +50 -0
  2. eval_results/global_step_0/mmlu_stem/test_qwen-boxed_-1_seed0_t1.0_s0_e-1.jsonl +3 -0
  3. eval_results/plots/eval_results_acc_keywords.png +3 -0
  4. eval_results/plots/eval_results_acc_pass_acc.png +3 -0
  5. eval_results/plots/eval_results_acc_tokens.png +3 -0
  6. eval_results/plots/eval_results_avg_stop_tokens.png +3 -0
  7. eval_results/plots/eval_results_box_ratio_and_token_length.png +3 -0
  8. eval_results/plots/eval_results_clip_ratio.png +3 -0
  9. eval_results/plots/eval_results_correct_tokens.png +3 -0
  10. eval_results/plots/eval_results_repeat_ratio_and_token_length.png +3 -0
  11. eval_results/plots/eval_results_tokens_keywords.png +3 -0
  12. eval_results/plots/eval_results_wrong_tokens.png +3 -0
  13. eval_results_avg16/plots/eval_results_avg16_repeat_ratio_and_token_length.png +3 -0
  14. eval_results_avg16/plots/eval_results_avg16_wrong_tokens.png +3 -0
  15. eval_results_avg32/plots/eval_results_avg32_acc_keywords.png +3 -0
  16. eval_results_avg32/plots/eval_results_avg32_acc_pass_acc.png +3 -0
  17. eval_results_avg32/plots/eval_results_avg32_acc_tokens.png +3 -0
  18. eval_results_avg32/plots/eval_results_avg32_avg_stop_tokens.png +3 -0
  19. eval_results_avg32/plots/eval_results_avg32_box_ratio_and_token_length.png +3 -0
  20. eval_results_avg32/plots/eval_results_avg32_clip_ratio.png +3 -0
  21. eval_results_avg32/plots/eval_results_avg32_correct_tokens.png +3 -0
  22. eval_results_avg32/plots/eval_results_avg32_repeat_ratio_and_token_length.png +3 -0
  23. eval_results_avg32/plots/eval_results_avg32_tokens_keywords.png +3 -0
  24. eval_results_avg32/plots/eval_results_avg32_wrong_tokens.png +3 -0
  25. eval_results_avg4/plots/eval_results_avg4_acc_keywords.png +3 -0
  26. eval_results_avg4/plots/eval_results_avg4_acc_pass_acc.png +3 -0
  27. eval_results_avg4/plots/eval_results_avg4_acc_tokens.png +3 -0
  28. eval_results_avg4/plots/eval_results_avg4_avg_stop_tokens.png +3 -0
  29. eval_results_avg4/plots/eval_results_avg4_box_ratio_and_token_length.png +3 -0
  30. eval_results_avg4/plots/eval_results_avg4_clip_ratio.png +3 -0
  31. eval_results_avg4/plots/eval_results_avg4_correct_tokens.png +3 -0
  32. eval_results_avg4/plots/eval_results_avg4_repeat_ratio_and_token_length.png +3 -0
  33. eval_results_avg4/plots/eval_results_avg4_tokens_keywords.png +3 -0
  34. eval_results_avg4/plots/eval_results_avg4_wrong_tokens.png +3 -0
  35. eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_extended|ifeval|0_2025-08-22T21-59-53.252655.csv +3 -0
  36. eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_extended|ifeval|0_2025-08-22T21-59-53.252655.parquet +3 -0
  37. eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_lighteval|gpqa:diamond|0_2025-08-22T21-59-53.252655.parquet +3 -0
  38. eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-24T06-10-51.524843/details_extended|ifeval|0_2025-08-24T06-10-51.524843.csv +3 -0
  39. eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-24T06-10-51.524843/details_extended|ifeval|0_2025-08-24T06-10-51.524843.parquet +3 -0
  40. eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_extended|ifeval|0_2025-08-22T21-51-40.106242.csv +3 -0
  41. eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_extended|ifeval|0_2025-08-22T21-51-40.106242.parquet +3 -0
  42. eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_lighteval|gpqa:diamond|0_2025-08-22T21-51-40.106242.parquet +3 -0
  43. eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-24T06-01-33.728040/details_extended|ifeval|0_2025-08-24T06-01-33.728040.csv +3 -0
  44. eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-24T06-01-33.728040/details_extended|ifeval|0_2025-08-24T06-01-33.728040.parquet +3 -0
  45. eval_results_ood/global_step_100/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_100_actor_huggingface/2025-08-22T22-32-53.893348/details_extended|ifeval|0_2025-08-22T22-32-53.893348.parquet +3 -0
  46. eval_results_ood/global_step_100/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_100_actor_huggingface/2025-08-22T22-32-53.893348/details_lighteval|gpqa:diamond|0_2025-08-22T22-32-53.893348.parquet +3 -0
  47. eval_results_ood/global_step_100/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_100_actor_huggingface/2025-08-24T06-36-23.774981/details_extended|ifeval|0_2025-08-24T06-36-23.774981.parquet +3 -0
  48. eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_extended|ifeval|0_2025-08-22T21-50-15.704473.csv +3 -0
  49. eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_extended|ifeval|0_2025-08-22T21-50-15.704473.parquet +3 -0
  50. eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_lighteval|gpqa:diamond|0_2025-08-22T21-50-15.704473.parquet +3 -0
.gitattributes CHANGED
@@ -43,3 +43,53 @@ global_step_10/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -t
43
  global_step_100/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
  global_step_20/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
45
  global_step_30/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
43
  global_step_100/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
  global_step_20/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
45
  global_step_30/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
46
+ eval_results/global_step_0/mmlu_stem/test_qwen-boxed_-1_seed0_t1.0_s0_e-1.jsonl filter=lfs diff=lfs merge=lfs -text
47
+ eval_results/plots/eval_results_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
48
+ eval_results/plots/eval_results_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
49
+ eval_results/plots/eval_results_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
50
+ eval_results/plots/eval_results_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
51
+ eval_results/plots/eval_results_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
52
+ eval_results/plots/eval_results_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
53
+ eval_results/plots/eval_results_repeat_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
54
+ eval_results/plots/eval_results_correct_tokens.png filter=lfs diff=lfs merge=lfs -text
55
+ eval_results/plots/eval_results_tokens_keywords.png filter=lfs diff=lfs merge=lfs -text
56
+ eval_results/plots/eval_results_wrong_tokens.png filter=lfs diff=lfs merge=lfs -text
57
+ eval_results_avg16/plots/eval_results_avg16_repeat_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
58
+ eval_results_avg16/plots/eval_results_avg16_wrong_tokens.png filter=lfs diff=lfs merge=lfs -text
59
+ eval_results_avg32/plots/eval_results_avg32_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
60
+ eval_results_avg32/plots/eval_results_avg32_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
61
+ eval_results_avg32/plots/eval_results_avg32_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
62
+ eval_results_avg32/plots/eval_results_avg32_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
63
+ eval_results_avg32/plots/eval_results_avg32_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
64
+ eval_results_avg32/plots/eval_results_avg32_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
65
+ eval_results_avg32/plots/eval_results_avg32_correct_tokens.png filter=lfs diff=lfs merge=lfs -text
66
+ eval_results_avg32/plots/eval_results_avg32_tokens_keywords.png filter=lfs diff=lfs merge=lfs -text
67
+ eval_results_avg32/plots/eval_results_avg32_repeat_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
68
+ eval_results_avg32/plots/eval_results_avg32_wrong_tokens.png filter=lfs diff=lfs merge=lfs -text
69
+ eval_results_avg4/plots/eval_results_avg4_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
70
+ eval_results_avg4/plots/eval_results_avg4_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
71
+ eval_results_avg4/plots/eval_results_avg4_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
72
+ eval_results_avg4/plots/eval_results_avg4_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
73
+ eval_results_avg4/plots/eval_results_avg4_correct_tokens.png filter=lfs diff=lfs merge=lfs -text
74
+ eval_results_avg4/plots/eval_results_avg4_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
75
+ eval_results_avg4/plots/eval_results_avg4_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
76
+ eval_results_avg4/plots/eval_results_avg4_repeat_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
77
+ eval_results_avg4/plots/eval_results_avg4_tokens_keywords.png filter=lfs diff=lfs merge=lfs -text
78
+ eval_results_avg4/plots/eval_results_avg4_wrong_tokens.png filter=lfs diff=lfs merge=lfs -text
79
+ eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-24T06-10-51.524843/details_extended|ifeval|0_2025-08-24T06-10-51.524843.csv filter=lfs diff=lfs merge=lfs -text
80
+ eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_extended|ifeval|0_2025-08-22T21-59-53.252655.csv filter=lfs diff=lfs merge=lfs -text
81
+ eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_extended|ifeval|0_2025-08-22T21-51-40.106242.csv filter=lfs diff=lfs merge=lfs -text
82
+ eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-24T06-01-33.728040/details_extended|ifeval|0_2025-08-24T06-01-33.728040.csv filter=lfs diff=lfs merge=lfs -text
83
+ eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-24T05-55-03.541337/details_extended|ifeval|0_2025-08-24T05-55-03.541337.csv filter=lfs diff=lfs merge=lfs -text
84
+ eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_extended|ifeval|0_2025-08-22T21-50-15.704473.csv filter=lfs diff=lfs merge=lfs -text
85
+ eval_results_ood/global_step_30/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_30_actor_huggingface/2025-08-24T05-44-37.776773/details_extended|ifeval|0_2025-08-24T05-44-37.776773.csv filter=lfs diff=lfs merge=lfs -text
86
+ eval_results_ood/global_step_30/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_30_actor_huggingface/2025-08-22T21-38-21.345863/details_extended|ifeval|0_2025-08-22T21-38-21.345863.csv filter=lfs diff=lfs merge=lfs -text
87
+ eval_results_ood/global_step_40/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_40_actor_huggingface/2025-08-22T21-39-31.356295/details_extended|ifeval|0_2025-08-22T21-39-31.356295.csv filter=lfs diff=lfs merge=lfs -text
88
+ eval_results_ood/global_step_50/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_50_actor_huggingface/2025-08-22T21-39-22.119596/details_extended|ifeval|0_2025-08-22T21-39-22.119596.csv filter=lfs diff=lfs merge=lfs -text
89
+ eval_results_ood/global_step_50/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_50_actor_huggingface/2025-08-24T05-51-56.876297/details_extended|ifeval|0_2025-08-24T05-51-56.876297.csv filter=lfs diff=lfs merge=lfs -text
90
+ eval_results_ood/global_step_40/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_40_actor_huggingface/2025-08-24T05-54-37.431353/details_extended|ifeval|0_2025-08-24T05-54-37.431353.csv filter=lfs diff=lfs merge=lfs -text
91
+ eval_results_ood/global_step_60/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_60_actor_huggingface/2025-08-22T21-37-55.750769/details_extended|ifeval|0_2025-08-22T21-37-55.750769.csv filter=lfs diff=lfs merge=lfs -text
92
+ eval_results_ood/global_step_70/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_70_actor_huggingface/2025-08-22T22-37-40.349268/details_extended|ifeval|0_2025-08-22T22-37-40.349268.csv filter=lfs diff=lfs merge=lfs -text
93
+ eval_results_ood/global_step_70/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_70_actor_huggingface/2025-08-24T05-44-33.516844/details_extended|ifeval|0_2025-08-24T05-44-33.516844.csv filter=lfs diff=lfs merge=lfs -text
94
+ eval_results_ood/global_step_80/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_80_actor_huggingface/2025-08-22T22-42-52.264172/details_extended|ifeval|0_2025-08-22T22-42-52.264172.csv filter=lfs diff=lfs merge=lfs -text
95
+ eval_results_ood/global_step_90/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_90_actor_huggingface/2025-08-24T06-44-12.807078/details_extended|ifeval|0_2025-08-24T06-44-12.807078.csv filter=lfs diff=lfs merge=lfs -text
eval_results/global_step_0/mmlu_stem/test_qwen-boxed_-1_seed0_t1.0_s0_e-1.jsonl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ff9273f7375822df00a8822d8d2baf9d15e646641ddca406e49d399f41b0a3b
3
+ size 11883083
eval_results/plots/eval_results_acc_keywords.png ADDED

Git LFS Details

  • SHA256: 49d1656a55a0662e5a20c4e4741770c26a920957052955fbcf4040d980287c21
  • Pointer size: 131 Bytes
  • Size of remote file: 292 kB
eval_results/plots/eval_results_acc_pass_acc.png ADDED

Git LFS Details

  • SHA256: a2039181e50075a3083e9a0742a3402ce26091fce58b3b4c70a00b7faa790c82
  • Pointer size: 131 Bytes
  • Size of remote file: 215 kB
eval_results/plots/eval_results_acc_tokens.png ADDED

Git LFS Details

  • SHA256: 062a9050fcaaddc71d6b3d3143d7cfbf6bf7f4dd0b86deedac81c1eeacddbeaf
  • Pointer size: 131 Bytes
  • Size of remote file: 283 kB
eval_results/plots/eval_results_avg_stop_tokens.png ADDED

Git LFS Details

  • SHA256: 5928e4d2dfed9aa37efbbdc5bd8aeef4b73206176d87d5b6475dae2e3501b8fa
  • Pointer size: 131 Bytes
  • Size of remote file: 315 kB
eval_results/plots/eval_results_box_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: e840c11d4a0a5e0d0a417f3bc1cb387873fdb72ce1689b25ccf3f9b1564f1c84
  • Pointer size: 131 Bytes
  • Size of remote file: 280 kB
eval_results/plots/eval_results_clip_ratio.png ADDED

Git LFS Details

  • SHA256: 559a8b3dfbe72f9dbba7491ba044704710e530ec56155fd406a76289eb1005c9
  • Pointer size: 131 Bytes
  • Size of remote file: 283 kB
eval_results/plots/eval_results_correct_tokens.png ADDED

Git LFS Details

  • SHA256: 00ac3a0fe0b8e30eb8a3bd44b5242334877c9fd9238a989d16292470768d37de
  • Pointer size: 131 Bytes
  • Size of remote file: 287 kB
eval_results/plots/eval_results_repeat_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: 8924af34cff658c3e55191206d4405a0e0a0a11e431c8ec427537d529049c8da
  • Pointer size: 131 Bytes
  • Size of remote file: 310 kB
eval_results/plots/eval_results_tokens_keywords.png ADDED

Git LFS Details

  • SHA256: 39d5974275f2903e0c0711af9dbe0320bf0c00b77ecf47ad0711234d69e4c211
  • Pointer size: 131 Bytes
  • Size of remote file: 293 kB
eval_results/plots/eval_results_wrong_tokens.png ADDED

Git LFS Details

  • SHA256: 36a6603ad901da0194e806833c7fdfecb100c61a606ed2cc02a342c932060b23
  • Pointer size: 131 Bytes
  • Size of remote file: 294 kB
eval_results_avg16/plots/eval_results_avg16_repeat_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: d738dad7c3ca245fd23637954ad624278714fe3c7730c0136041246702fea45f
  • Pointer size: 131 Bytes
  • Size of remote file: 102 kB
eval_results_avg16/plots/eval_results_avg16_wrong_tokens.png ADDED

Git LFS Details

  • SHA256: 36baffd9219c6b7e13ce8866239fa695ea67ba57638bf066d182a2b00e7c544e
  • Pointer size: 131 Bytes
  • Size of remote file: 100 kB
eval_results_avg32/plots/eval_results_avg32_acc_keywords.png ADDED

Git LFS Details

  • SHA256: ffe019ecf18789938f2914af5befc3d297bce915f38f351f446e58527e1f8ad7
  • Pointer size: 131 Bytes
  • Size of remote file: 124 kB
eval_results_avg32/plots/eval_results_avg32_acc_pass_acc.png ADDED

Git LFS Details

  • SHA256: 1daff32874b1a4490d66d4f54a9ab62064a6d5d7737294b77d056a7128c2ec0a
  • Pointer size: 131 Bytes
  • Size of remote file: 132 kB
eval_results_avg32/plots/eval_results_avg32_acc_tokens.png ADDED

Git LFS Details

  • SHA256: 0957cac82db2eab5efa9ddc4e34ff919bc87bdc0a3f96eddc9aede900fb8e607
  • Pointer size: 131 Bytes
  • Size of remote file: 152 kB
eval_results_avg32/plots/eval_results_avg32_avg_stop_tokens.png ADDED

Git LFS Details

  • SHA256: d69abe512b09a0abab3e6e391238478bd0ff3428f5e87cb21dc230a00bb23792
  • Pointer size: 131 Bytes
  • Size of remote file: 158 kB
eval_results_avg32/plots/eval_results_avg32_box_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: 13d6743269ab15baffd40a4f8cbe5767f2fc54b93fee8520ed3db0a6aeb92853
  • Pointer size: 131 Bytes
  • Size of remote file: 176 kB
eval_results_avg32/plots/eval_results_avg32_clip_ratio.png ADDED

Git LFS Details

  • SHA256: 8a9f7ca840a302c9b50c2b56c659a48ba7f959cdfecbd024b6cb3de72f8427d6
  • Pointer size: 131 Bytes
  • Size of remote file: 137 kB
eval_results_avg32/plots/eval_results_avg32_correct_tokens.png ADDED

Git LFS Details

  • SHA256: 2a44607a42dcc83b19c8b13a97475ca7086e820dd5afd9ba91e8df40f1a20177
  • Pointer size: 131 Bytes
  • Size of remote file: 142 kB
eval_results_avg32/plots/eval_results_avg32_repeat_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: 1c878a941fd573e577b584f1d8d91366d2289d2b688a236a2add55fb9d08288a
  • Pointer size: 131 Bytes
  • Size of remote file: 170 kB
eval_results_avg32/plots/eval_results_avg32_tokens_keywords.png ADDED

Git LFS Details

  • SHA256: 96e704968b803ad96e0e4fb59a30e6ac964300105a44145d5f677c74f6eac32d
  • Pointer size: 131 Bytes
  • Size of remote file: 144 kB
eval_results_avg32/plots/eval_results_avg32_wrong_tokens.png ADDED

Git LFS Details

  • SHA256: cd23b750d36fa07f8b19c21a1205e290d2f1b976faf20a50e25178fa5259e479
  • Pointer size: 131 Bytes
  • Size of remote file: 152 kB
eval_results_avg4/plots/eval_results_avg4_acc_keywords.png ADDED

Git LFS Details

  • SHA256: 8226f30d0d489d34bfa0f0da43a963b3b4d97eaaeb7c1ea4d618a9d37b8b0403
  • Pointer size: 131 Bytes
  • Size of remote file: 213 kB
eval_results_avg4/plots/eval_results_avg4_acc_pass_acc.png ADDED

Git LFS Details

  • SHA256: 9624544ced7837efc1ccf28590313e83742d731d6988f5c8b5e674094d66d6a0
  • Pointer size: 131 Bytes
  • Size of remote file: 197 kB
eval_results_avg4/plots/eval_results_avg4_acc_tokens.png ADDED

Git LFS Details

  • SHA256: a9bdc548ccf077a831e8eb19be60e6e1145276b9731074ffe8e6b5227a2a6ce3
  • Pointer size: 131 Bytes
  • Size of remote file: 221 kB
eval_results_avg4/plots/eval_results_avg4_avg_stop_tokens.png ADDED

Git LFS Details

  • SHA256: fdfbc262c1f9d499ca5d541813ea2c716277f38e8a1a876bb5e4ed1e3134e653
  • Pointer size: 131 Bytes
  • Size of remote file: 247 kB
eval_results_avg4/plots/eval_results_avg4_box_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: 0dbbdf379e7fee0bb289e10b0af5827658d6dc9f62cf6ad96ccfc63637a2ecd3
  • Pointer size: 131 Bytes
  • Size of remote file: 236 kB
eval_results_avg4/plots/eval_results_avg4_clip_ratio.png ADDED

Git LFS Details

  • SHA256: be3d7ead070142ee1b6993c89aa562f4464b0ab8a5ad256e49b2db77a712d588
  • Pointer size: 131 Bytes
  • Size of remote file: 212 kB
eval_results_avg4/plots/eval_results_avg4_correct_tokens.png ADDED

Git LFS Details

  • SHA256: 5f32c43f35e915847fe8508fa23eeb83796b3bc3822a7f1a0fc5792a9e29ecf1
  • Pointer size: 131 Bytes
  • Size of remote file: 216 kB
eval_results_avg4/plots/eval_results_avg4_repeat_ratio_and_token_length.png ADDED

Git LFS Details

  • SHA256: 3b2c7b65309199cd4bc0ec93831883d3669a30b7ab7dcfe741f253c0d4e653ba
  • Pointer size: 131 Bytes
  • Size of remote file: 257 kB
eval_results_avg4/plots/eval_results_avg4_tokens_keywords.png ADDED

Git LFS Details

  • SHA256: 62cd87f213d722301098e490aeb2fc543a20e12b19f94d87414a58db30625460
  • Pointer size: 131 Bytes
  • Size of remote file: 231 kB
eval_results_avg4/plots/eval_results_avg4_wrong_tokens.png ADDED

Git LFS Details

  • SHA256: 80c25342d27a5c8a5a37751ab51e271a7c2c5008b4e3fcdc35a1d1b06bb99f6b
  • Pointer size: 131 Bytes
  • Size of remote file: 217 kB
eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_extended|ifeval|0_2025-08-22T21-59-53.252655.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9800b44e313740bb47d6b2173f893b1a920e59b74224c8fb2c11fa4b2fcc2d6a
3
+ size 19121550
eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_extended|ifeval|0_2025-08-22T21-59-53.252655.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9593b94d437284733b033513af89f919c25eb4382eea1d3904bf70ce47596ca
3
+ size 2883389
eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-22T21-59-53.252655/details_lighteval|gpqa:diamond|0_2025-08-22T21-59-53.252655.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be1ed464355dc0c52d62483bc62dcd9195068359dafbe13fe6a9969ec92abea9
3
+ size 496418
eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-24T06-10-51.524843/details_extended|ifeval|0_2025-08-24T06-10-51.524843.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d20a05df2ee2bd81ee63cda2f85422fbcd6ce4865abe8d05340d819665446067
3
+ size 19994943
eval_results_ood/global_step_0/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_0_actor_huggingface/2025-08-24T06-10-51.524843/details_extended|ifeval|0_2025-08-24T06-10-51.524843.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d9bbb67c4a575fd02a483cb86d8c85faa77383b20135fc673cb238b67cd1249
3
+ size 3245950
eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_extended|ifeval|0_2025-08-22T21-51-40.106242.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b55eef8612ff417e53075228c7f898ab2665e6211a81752a38b1449a69d493b8
3
+ size 18066599
eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_extended|ifeval|0_2025-08-22T21-51-40.106242.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67918ede7592f1ec0f12a907f018975edb932b7da888785103ff31e25e5c6b80
3
+ size 2783976
eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-22T21-51-40.106242/details_lighteval|gpqa:diamond|0_2025-08-22T21-51-40.106242.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69aeee0323f4544c38630ee5a8383944c6867782144ab14333b22895a5c27d41
3
+ size 600158
eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-24T06-01-33.728040/details_extended|ifeval|0_2025-08-24T06-01-33.728040.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff26a8d3aed192b4c07acfe742ec88acf199702fec683282964badf1883baea3
3
+ size 17528529
eval_results_ood/global_step_10/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_10_actor_huggingface/2025-08-24T06-01-33.728040/details_extended|ifeval|0_2025-08-24T06-01-33.728040.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b1bb209088f341de8656bc4293bd967e80b37e0fe87122af5850c00cc03e7e7
3
+ size 2976719
eval_results_ood/global_step_100/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_100_actor_huggingface/2025-08-22T22-32-53.893348/details_extended|ifeval|0_2025-08-22T22-32-53.893348.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3360d7a2b39ef6455a5efa3d06de2002566c852186f99fbefe18e4c92078180e
3
+ size 1491028
eval_results_ood/global_step_100/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_100_actor_huggingface/2025-08-22T22-32-53.893348/details_lighteval|gpqa:diamond|0_2025-08-22T22-32-53.893348.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31248d7d49794a2e63f7f59128354fdaea3e4ab35f82e3f0fbc30ccf605b8391
3
+ size 531581
eval_results_ood/global_step_100/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_100_actor_huggingface/2025-08-24T06-36-23.774981/details_extended|ifeval|0_2025-08-24T06-36-23.774981.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c86d7e783cd257a33d71aed2f889b0600c538aa7a76a9f2f31bb9febd4772b6
3
+ size 1520595
eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_extended|ifeval|0_2025-08-22T21-50-15.704473.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9708b743e2cb74cd0a117aa3f771f71c46c4147928d745573c3f0e81821308d9
3
+ size 16977649
eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_extended|ifeval|0_2025-08-22T21-50-15.704473.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a469d26aac2e448511d3cce0294051e36f2065aff3c27adcd373cc3ac3c76bb
3
+ size 2387399
eval_results_ood/global_step_20/details/_home_work_minzijun_rl_output_2_checkpoints_verl_role_sft_grpo_deepmath_train_sample_6144_context_4k_Qwen3-1.7B-base_max_response4096_batch1024_rollout8_vllm_no_shuffle_global_step_20_actor_huggingface/2025-08-22T21-50-15.704473/details_lighteval|gpqa:diamond|0_2025-08-22T21-50-15.704473.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef0b54f3ae01ac98f2fe23157c71fc9c0562cc792fb83f1dfd5d5ade6fa5b0aa
3
+ size 627710