| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9893390191897654, | |
| "eval_steps": 100, | |
| "global_step": 58, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "completion_length": 616.8067230224609, | |
| "epoch": 0.08528784648187633, | |
| "grad_norm": 3.238640308380127, | |
| "kl": 0.00016660690307617186, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0, | |
| "reward": 0.6430803842842578, | |
| "reward_std": 0.3440178133547306, | |
| "rewards/accuracy_reward": 0.6424107439815998, | |
| "rewards/format_reward": 0.0006696428870782256, | |
| "step": 5 | |
| }, | |
| { | |
| "completion_length": 612.4875190734863, | |
| "epoch": 0.17057569296375266, | |
| "grad_norm": 0.19049902260303497, | |
| "kl": 0.0018734931945800781, | |
| "learning_rate": 2.956412726139078e-06, | |
| "loss": 0.0001, | |
| "reward": 0.7142857521772384, | |
| "reward_std": 0.2628630593419075, | |
| "rewards/accuracy_reward": 0.7142857521772384, | |
| "rewards/format_reward": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 621.501594543457, | |
| "epoch": 0.255863539445629, | |
| "grad_norm": 4.2737298011779785, | |
| "kl": 1.1404457092285156, | |
| "learning_rate": 2.7836719084521715e-06, | |
| "loss": 0.0456, | |
| "reward": 0.7477678894996643, | |
| "reward_std": 0.23944801036268473, | |
| "rewards/accuracy_reward": 0.7477678894996643, | |
| "rewards/format_reward": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "completion_length": 609.3560501098633, | |
| "epoch": 0.3411513859275053, | |
| "grad_norm": 0.15062430500984192, | |
| "kl": 0.007084274291992187, | |
| "learning_rate": 2.4946839873611927e-06, | |
| "loss": 0.0003, | |
| "reward": 0.7607143253087998, | |
| "reward_std": 0.22700632140040397, | |
| "rewards/accuracy_reward": 0.7607143253087998, | |
| "rewards/format_reward": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "completion_length": 614.4294906616211, | |
| "epoch": 0.42643923240938164, | |
| "grad_norm": 1.7865530252456665, | |
| "kl": 0.005679512023925781, | |
| "learning_rate": 2.1156192081791355e-06, | |
| "loss": 0.0002, | |
| "reward": 0.756473246216774, | |
| "reward_std": 0.1937894519418478, | |
| "rewards/accuracy_reward": 0.756473246216774, | |
| "rewards/format_reward": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "completion_length": 614.8489112854004, | |
| "epoch": 0.511727078891258, | |
| "grad_norm": 0.26249098777770996, | |
| "kl": 0.0034336090087890626, | |
| "learning_rate": 1.6808050203829845e-06, | |
| "loss": 0.0001, | |
| "reward": 0.7555803924798965, | |
| "reward_std": 0.1806625969707966, | |
| "rewards/accuracy_reward": 0.7555803924798965, | |
| "rewards/format_reward": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "completion_length": 604.0652038574219, | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 0.15777133405208588, | |
| "kl": 0.0031137466430664062, | |
| "learning_rate": 1.2296174432791415e-06, | |
| "loss": 0.0001, | |
| "reward": 0.7430803917348385, | |
| "reward_std": 0.1779588583856821, | |
| "rewards/accuracy_reward": 0.7430803917348385, | |
| "rewards/format_reward": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "completion_length": 589.3364120483399, | |
| "epoch": 0.6823027718550106, | |
| "grad_norm": 0.14344827830791473, | |
| "kl": 0.003224945068359375, | |
| "learning_rate": 8.029152419343472e-07, | |
| "loss": 0.0001, | |
| "reward": 0.7691964641213417, | |
| "reward_std": 0.16417927835136653, | |
| "rewards/accuracy_reward": 0.7691964641213417, | |
| "rewards/format_reward": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "completion_length": 607.6384132385253, | |
| "epoch": 0.767590618336887, | |
| "grad_norm": 0.0754118412733078, | |
| "kl": 0.0033246994018554686, | |
| "learning_rate": 4.3933982822017883e-07, | |
| "loss": 0.0001, | |
| "reward": 0.7488839641213417, | |
| "reward_std": 0.18731108466163277, | |
| "rewards/accuracy_reward": 0.7488839641213417, | |
| "rewards/format_reward": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "completion_length": 608.0522613525391, | |
| "epoch": 0.8528784648187633, | |
| "grad_norm": 0.07660669833421707, | |
| "kl": 0.003369140625, | |
| "learning_rate": 1.718159615201853e-07, | |
| "loss": 0.0001, | |
| "reward": 0.7475446820259094, | |
| "reward_std": 0.18219188433140515, | |
| "rewards/accuracy_reward": 0.7475446820259094, | |
| "rewards/format_reward": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "completion_length": 604.2169937133789, | |
| "epoch": 0.9381663113006397, | |
| "grad_norm": 0.19257374107837677, | |
| "kl": 0.003157806396484375, | |
| "learning_rate": 2.4570139579284723e-08, | |
| "loss": 0.0001, | |
| "reward": 0.7812500342726707, | |
| "reward_std": 0.18055678810924292, | |
| "rewards/accuracy_reward": 0.7812500342726707, | |
| "rewards/format_reward": 0.0, | |
| "step": 55 | |
| }, | |
| { | |
| "completion_length": 595.3329900105795, | |
| "epoch": 0.9893390191897654, | |
| "kl": 0.0034383138020833335, | |
| "reward": 0.7645089626312256, | |
| "reward_std": 0.16559721545005837, | |
| "rewards/accuracy_reward": 0.7645089626312256, | |
| "rewards/format_reward": 0.0, | |
| "step": 58, | |
| "total_flos": 0.0, | |
| "train_loss": 0.004416576032875535, | |
| "train_runtime": 7299.7333, | |
| "train_samples_per_second": 1.027, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 58, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |