| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9952755905511811, | |
| "eval_steps": 2000000, | |
| "global_step": 79, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 814.1228103637695, | |
| "epoch": 0.012598425196850394, | |
| "grad_norm": 0.4256682094610429, | |
| "learning_rate": 1.25e-07, | |
| "loss": 0.1859, | |
| "num_tokens": 857038.0, | |
| "reward": 0.5837053805589676, | |
| "reward_std": 0.38259728997945786, | |
| "rewards/accuracy_reward": 0.578125, | |
| "rewards/format_reward": 0.011160714784637094, | |
| "step": 1 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 799.2157211303711, | |
| "epoch": 0.06299212598425197, | |
| "grad_norm": 0.2682297103080829, | |
| "learning_rate": 6.249999999999999e-07, | |
| "loss": 0.1749, | |
| "num_tokens": 4268899.0, | |
| "reward": 0.5555245783179998, | |
| "reward_std": 0.3812973015010357, | |
| "rewards/accuracy_reward": 0.5532924123108387, | |
| "rewards/format_reward": 0.004464285884751007, | |
| "step": 5 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 818.2045013427735, | |
| "epoch": 0.12598425196850394, | |
| "grad_norm": 0.3636912627924593, | |
| "learning_rate": 9.980434110374724e-07, | |
| "loss": 0.163, | |
| "num_tokens": 8619311.0, | |
| "reward": 0.5619419872760772, | |
| "reward_std": 0.3748646147549152, | |
| "rewards/accuracy_reward": 0.5591517865657807, | |
| "rewards/format_reward": 0.005580357369035482, | |
| "step": 10 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 825.4431121826171, | |
| "epoch": 0.1889763779527559, | |
| "grad_norm": 0.5533896497151992, | |
| "learning_rate": 9.762072666790656e-07, | |
| "loss": 0.1654, | |
| "num_tokens": 12961552.0, | |
| "reward": 0.5640625230967998, | |
| "reward_std": 0.36274583265185356, | |
| "rewards/accuracy_reward": 0.5607142888009549, | |
| "rewards/format_reward": 0.006696428847499191, | |
| "step": 15 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 836.8739181518555, | |
| "epoch": 0.25196850393700787, | |
| "grad_norm": 0.23700630720254828, | |
| "learning_rate": 9.311572862600138e-07, | |
| "loss": 0.1757, | |
| "num_tokens": 17365979.0, | |
| "reward": 0.5938616335391999, | |
| "reward_std": 0.3463574357330799, | |
| "rewards/accuracy_reward": 0.5930803567171097, | |
| "rewards/format_reward": 0.0015625000698491931, | |
| "step": 20 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 818.7475799560547, | |
| "epoch": 0.31496062992125984, | |
| "grad_norm": 0.5349999125528991, | |
| "learning_rate": 8.650895363529172e-07, | |
| "loss": 0.1813, | |
| "num_tokens": 21690080.0, | |
| "reward": 0.6300223484635353, | |
| "reward_std": 0.3288935709744692, | |
| "rewards/accuracy_reward": 0.630769232660532, | |
| "rewards/format_reward": 0.0011160714784637094, | |
| "step": 25 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 779.5919998168945, | |
| "epoch": 0.3779527559055118, | |
| "grad_norm": 0.33764515237713627, | |
| "learning_rate": 7.812246438203903e-07, | |
| "loss": 0.1706, | |
| "num_tokens": 25851452.0, | |
| "reward": 0.6796875342726707, | |
| "reward_std": 0.28653539419174195, | |
| "rewards/accuracy_reward": 0.6794642806053162, | |
| "rewards/format_reward": 0.00044642859138548373, | |
| "step": 30 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 764.461865234375, | |
| "epoch": 0.4409448818897638, | |
| "grad_norm": 0.17429005608156234, | |
| "learning_rate": 6.836507988323784e-07, | |
| "loss": 0.1559, | |
| "num_tokens": 29947561.0, | |
| "reward": 0.6868303909897804, | |
| "reward_std": 0.27214415185153484, | |
| "rewards/accuracy_reward": 0.6866071395576, | |
| "rewards/format_reward": 0.00044642859138548373, | |
| "step": 35 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 762.9011520385742, | |
| "epoch": 0.5039370078740157, | |
| "grad_norm": 0.14957998644439033, | |
| "learning_rate": 5.771244664826511e-07, | |
| "loss": 0.1182, | |
| "num_tokens": 34064990.0, | |
| "reward": 0.6814732447266578, | |
| "reward_std": 0.2671396616846323, | |
| "rewards/accuracy_reward": 0.6814732149243354, | |
| "rewards/format_reward": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 744.8989166259765, | |
| "epoch": 0.5669291338582677, | |
| "grad_norm": 0.7244950959434313, | |
| "learning_rate": 4.6683852178244817e-07, | |
| "loss": 0.1165, | |
| "num_tokens": 38094369.0, | |
| "reward": 0.6755580686032772, | |
| "reward_std": 0.26786904707551, | |
| "rewards/accuracy_reward": 0.6754464261233807, | |
| "rewards/format_reward": 0.00022321429569274187, | |
| "step": 45 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 758.1707931518555, | |
| "epoch": 0.6299212598425197, | |
| "grad_norm": 0.12379763987674317, | |
| "learning_rate": 3.5816911083285164e-07, | |
| "loss": 0.1091, | |
| "num_tokens": 42164678.0, | |
| "reward": 0.6937500357627868, | |
| "reward_std": 0.2522166069597006, | |
| "rewards/accuracy_reward": 0.6937499977648258, | |
| "rewards/format_reward": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 750.6433380126953, | |
| "epoch": 0.6929133858267716, | |
| "grad_norm": 0.22358072329593362, | |
| "learning_rate": 2.5641357801960184e-07, | |
| "loss": 0.1096, | |
| "num_tokens": 46185264.0, | |
| "reward": 0.6895089626312256, | |
| "reward_std": 0.2542113933712244, | |
| "rewards/accuracy_reward": 0.6895089261233807, | |
| "rewards/format_reward": 0.0, | |
| "step": 55 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 730.3611877441406, | |
| "epoch": 0.7559055118110236, | |
| "grad_norm": 0.42247506921603206, | |
| "learning_rate": 1.665322345816746e-07, | |
| "loss": 0.0853, | |
| "num_tokens": 50096882.0, | |
| "reward": 0.7053571708500386, | |
| "reward_std": 0.23149009980261326, | |
| "rewards/accuracy_reward": 0.7053571432828903, | |
| "rewards/format_reward": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 739.0698974609375, | |
| "epoch": 0.8188976377952756, | |
| "grad_norm": 0.33241630532899313, | |
| "learning_rate": 9.290655664821296e-08, | |
| "loss": 0.0931, | |
| "num_tokens": 54062275.0, | |
| "reward": 0.7138393133878708, | |
| "reward_std": 0.23870391435921193, | |
| "rewards/accuracy_reward": 0.7138392895460128, | |
| "rewards/format_reward": 0.0, | |
| "step": 65 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 713.727931213379, | |
| "epoch": 0.8818897637795275, | |
| "grad_norm": 0.2063115196222285, | |
| "learning_rate": 3.912559994556086e-08, | |
| "loss": 0.0924, | |
| "num_tokens": 57926568.0, | |
| "reward": 0.7283482469618321, | |
| "reward_std": 0.23798817545175552, | |
| "rewards/accuracy_reward": 0.7294986329972744, | |
| "rewards/format_reward": 0.0, | |
| "step": 70 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 740.9406539916993, | |
| "epoch": 0.9448818897637795, | |
| "grad_norm": 0.1952309372717915, | |
| "learning_rate": 7.811042888637209e-09, | |
| "loss": 0.0778, | |
| "num_tokens": 61914142.0, | |
| "reward": 0.690178605914116, | |
| "reward_std": 0.23276512399315835, | |
| "rewards/accuracy_reward": 0.6901785716414451, | |
| "rewards/format_reward": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "clip_ratio": 0.0, | |
| "completion_length": 768.2404909133911, | |
| "epoch": 0.9952755905511811, | |
| "num_tokens": 65183080.0, | |
| "reward": 0.6978236902505159, | |
| "reward_std": 0.24844580795615911, | |
| "rewards/accuracy_reward": 0.6978236641734838, | |
| "rewards/format_reward": 0.0, | |
| "step": 79, | |
| "total_flos": 0.0, | |
| "train_loss": 0.1304081463361088, | |
| "train_runtime": 17251.9813, | |
| "train_samples_per_second": 0.515, | |
| "train_steps_per_second": 0.005 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 79, | |
| "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": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |