| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "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": 386.53125, | |
| "epoch": 0.1, | |
| "grad_norm": 11.838155746459961, | |
| "kl": 0.0, | |
| "learning_rate": 5e-07, | |
| "loss": -0.0, | |
| "reward": 0.49732657615095377, | |
| "reward_std": 0.6507052630768158, | |
| "rewards/concensus_correctness_reward_func": 0.05999999865889549, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.19863907300168648, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.1136874994263053, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 289.15625, | |
| "epoch": 0.2, | |
| "grad_norm": 18.39859390258789, | |
| "kl": 0.0021505101231014123, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 0.19581820373423398, | |
| "reward_std": 0.4207594294275623, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.20109945279546082, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.0677812471985817, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 445.0625, | |
| "epoch": 0.3, | |
| "grad_norm": 15.898308753967285, | |
| "kl": 0.0011719617286871653, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 0.1751822994556278, | |
| "reward_std": 0.43424874570337124, | |
| "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": 0.16821355387219228, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.006968748988583684, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 325.625, | |
| "epoch": 0.4, | |
| "grad_norm": 155.9239959716797, | |
| "kl": 0.010154961106309202, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.45496489957440645, | |
| "reward_std": 0.7626551468565594, | |
| "rewards/concensus_correctness_reward_func": 0.11950000002980232, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.23943364236038178, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.028968748403713107, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 406.40625, | |
| "epoch": 0.5, | |
| "grad_norm": 16.150299072265625, | |
| "kl": 0.0016187584624276496, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.3459215434268117, | |
| "reward_std": 0.6205616659717634, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.15970279998145998, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.0012812484055757523, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 402.1875, | |
| "epoch": 0.6, | |
| "grad_norm": 12.852953910827637, | |
| "kl": 0.0021437756731756963, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 0.7064267341047525, | |
| "reward_std": 0.9020523639192106, | |
| "rewards/concensus_correctness_reward_func": 0.17631249874830246, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.2865517425816506, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.056062495801597834, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 304.875, | |
| "epoch": 0.7, | |
| "grad_norm": 7.3621931076049805, | |
| "kl": 0.002075214839351247, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 0.36785096395760775, | |
| "reward_std": 0.46947701514000073, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.1592259633471258, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.08362499932991341, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 288.5, | |
| "epoch": 0.8, | |
| "grad_norm": 21.77416229248047, | |
| "kl": 0.002049503083981108, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.5508579231100157, | |
| "reward_std": 0.6642407482140698, | |
| "rewards/concensus_correctness_reward_func": 0.06012500077486038, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.2362329208990559, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.12949999934062362, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 434.53125, | |
| "epoch": 0.9, | |
| "grad_norm": 18.984968185424805, | |
| "kl": 0.001922691324580228, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 0.36621046916843625, | |
| "reward_std": 0.6980189112837252, | |
| "rewards/concensus_correctness_reward_func": 0.12012500315904617, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.13721047036960954, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.016125003341585398, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 406.34375, | |
| "epoch": 1.0, | |
| "grad_norm": 5.100473880767822, | |
| "kl": 0.0013912481890656636, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.1821420846972615, | |
| "reward_std": 1.2628907886974048, | |
| "rewards/concensus_correctness_reward_func": 0.6661874987185001, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.3125, | |
| "rewards/question_recreation_reward_func": 0.20657960418611765, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.0031250016763806343, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 2.489093321855762e-06, | |
| "train_runtime": 278.934, | |
| "train_samples_per_second": 1.147, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |