| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.7272727272727275, | |
| "eval_steps": 500, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "completion_length": 389.9375, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.47265443205833435, | |
| "kl": 0.0013143350515747443, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 1.2336252145469189, | |
| "reward_std": 1.3198441192507744, | |
| "rewards/concensus_correctness_reward_func": 0.0625, | |
| "rewards/consensus_reward_func": 0.375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5032501667737961, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.29287500213831663, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 395.42857142857144, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.45607730746269226, | |
| "kl": 0.0019246864997382676, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 1.3128481251852853, | |
| "reward_std": 1.5038241403443473, | |
| "rewards/concensus_correctness_reward_func": 0.17299999935286386, | |
| "rewards/consensus_reward_func": 0.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.3277410151703017, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.05357142857142857, | |
| "rewards/xmlcount_reward_func": 0.25853571934359415, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 387.7857142857143, | |
| "epoch": 2.0, | |
| "grad_norm": 0.5184553861618042, | |
| "kl": 0.0015573270419346435, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 1.3654808487210954, | |
| "reward_std": 0.8350333869457245, | |
| "rewards/concensus_correctness_reward_func": 0.14285714285714285, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5702666001660484, | |
| "rewards/soft_format_reward_func": 0.017857142857142856, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.18807142547198705, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 422.59375, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 1.9705733060836792, | |
| "kl": 0.0016508261396666057, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.8875295454636216, | |
| "reward_std": 1.2564833909273148, | |
| "rewards/concensus_correctness_reward_func": 0.11524999886751175, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.45112330093979836, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.07115625031292439, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 422.2142857142857, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.47363153100013733, | |
| "kl": 0.0013743416140122072, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.6273614508765084, | |
| "reward_std": 0.689619990331786, | |
| "rewards/concensus_correctness_reward_func": 0.03571428571428571, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.27939715129988535, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.151535730276789, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 442.2857142857143, | |
| "epoch": 4.0, | |
| "grad_norm": 5.771443843841553, | |
| "kl": 0.0016621236739281034, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 1.2324650755950384, | |
| "reward_std": 1.6661317774227686, | |
| "rewards/concensus_correctness_reward_func": 0.16742857013429915, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5394650612558637, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.07914285787514277, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 462.8125, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.42445477843284607, | |
| "kl": 0.0016633963969070464, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 0.912763200700283, | |
| "reward_std": 1.3838151879608631, | |
| "rewards/concensus_correctness_reward_func": 0.08887499943375587, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4303256794810295, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.1279375022277236, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 457.57142857142856, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.49823829531669617, | |
| "kl": 0.0013963170016982726, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.7656691925866264, | |
| "reward_std": 0.6999560615846089, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.6487406364509037, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.11692856677940913, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 415.7857142857143, | |
| "epoch": 6.0, | |
| "grad_norm": 0.5699564218521118, | |
| "kl": 0.0015232375091207878, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 0.7318344861268997, | |
| "reward_std": 1.1388877247061049, | |
| "rewards/concensus_correctness_reward_func": 0.07142857142857142, | |
| "rewards/consensus_reward_func": 0.2857142857142857, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4585487289088113, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": -0.11957144098622459, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 413.0625, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.613993227481842, | |
| "kl": 0.0015577992162434384, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.0791360288858414, | |
| "reward_std": 1.1513150930404663, | |
| "rewards/concensus_correctness_reward_func": 0.0625, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4709485024213791, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.2331875106319785, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 1.452610661090148e-06, | |
| "train_runtime": 578.8344, | |
| "train_samples_per_second": 0.553, | |
| "train_steps_per_second": 0.035 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |