| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002948765204570586, | |
| "eval_steps": 500, | |
| "global_step": 12, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 576.9375, | |
| "epoch": 0.0002457304337142155, | |
| "grad_norm": 0.0, | |
| "kl": 0.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "num_tokens": 12535.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/reward_func": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 879.0625, | |
| "epoch": 0.000491460867428431, | |
| "grad_norm": 0.11151622953359844, | |
| "kl": 0.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0776, | |
| "num_tokens": 27624.0, | |
| "reward": 0.125, | |
| "reward_std": 0.2314550280570984, | |
| "rewards/reward_func": 0.125, | |
| "step": 2 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 923.75, | |
| "epoch": 0.0007371913011426465, | |
| "grad_norm": 1.7342617962764825e-06, | |
| "kl": 0.0003814697265625, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "num_tokens": 43892.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/reward_func": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 682.9375, | |
| "epoch": 0.000982921734856862, | |
| "grad_norm": 0.1605198642986145, | |
| "kl": 0.00042629241943359375, | |
| "learning_rate": 5e-07, | |
| "loss": -0.1736, | |
| "num_tokens": 56019.0, | |
| "reward": 0.1875, | |
| "reward_std": 0.408231720328331, | |
| "rewards/reward_func": 0.1875, | |
| "step": 4 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 729.5, | |
| "epoch": 0.0012286521685710775, | |
| "grad_norm": 0.15191784640809727, | |
| "kl": 0.00035190582275390625, | |
| "learning_rate": 5e-07, | |
| "loss": 0.1212, | |
| "num_tokens": 69587.0, | |
| "reward": 0.0625, | |
| "reward_std": 0.1767766922712326, | |
| "rewards/reward_func": 0.0625, | |
| "step": 5 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 680.3125, | |
| "epoch": 0.001474382602285293, | |
| "grad_norm": 0.00010216089929885942, | |
| "kl": 0.0011739730834960938, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "num_tokens": 82064.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/reward_func": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 723.1875, | |
| "epoch": 0.0017201130359995086, | |
| "grad_norm": 0.053916157377856225, | |
| "kl": 0.00029850006103515625, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0843, | |
| "num_tokens": 96387.0, | |
| "reward": 0.0625, | |
| "reward_std": 0.1767766922712326, | |
| "rewards/reward_func": 0.0625, | |
| "step": 7 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 646.1875, | |
| "epoch": 0.001965843469713724, | |
| "grad_norm": 0.16676206881832856, | |
| "kl": 0.0003566741943359375, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0941, | |
| "num_tokens": 108918.0, | |
| "reward": 0.25, | |
| "reward_std": 0.4355512708425522, | |
| "rewards/reward_func": 0.25, | |
| "step": 8 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 605.875, | |
| "epoch": 0.0022115739034279398, | |
| "grad_norm": 1.5988321121918276e-06, | |
| "kl": 0.0003833770751953125, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "num_tokens": 120428.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/reward_func": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 551.4375, | |
| "epoch": 0.002457304337142155, | |
| "grad_norm": 0.16315342243657902, | |
| "kl": 0.00034236907958984375, | |
| "learning_rate": 5e-07, | |
| "loss": -0.0335, | |
| "num_tokens": 130291.0, | |
| "reward": 0.0625, | |
| "reward_std": 0.1767766922712326, | |
| "rewards/reward_func": 0.0625, | |
| "step": 10 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 757.3125, | |
| "epoch": 0.0027030347708563705, | |
| "grad_norm": 1.5633163558139792e-06, | |
| "kl": 0.00029850006103515625, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "num_tokens": 149048.0, | |
| "reward": 0.0, | |
| "reward_std": 0.0, | |
| "rewards/reward_func": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 766.375, | |
| "epoch": 0.002948765204570586, | |
| "grad_norm": 0.08173214068073498, | |
| "kl": 0.00022268295288085938, | |
| "learning_rate": 5e-07, | |
| "loss": 0.1401, | |
| "num_tokens": 162550.0, | |
| "reward": 0.125, | |
| "reward_std": 0.2314550280570984, | |
| "rewards/reward_func": 0.125, | |
| "step": 12 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 4069, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 12, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |