| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.992, | |
| "eval_steps": 200, | |
| "global_step": 62, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016, | |
| "grad_norm": 33.26401254999367, | |
| "learning_rate": 7.142857142857142e-08, | |
| "logits/generated": -2.5860352516174316, | |
| "logits/real": -2.4361655712127686, | |
| "logps/generated": -109.8505859375, | |
| "logps/real": -99.55709838867188, | |
| "loss": 0.7803, | |
| "rewards/accuracies": 0.0, | |
| "rewards/generated": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/real": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 40.749062783665345, | |
| "learning_rate": 4.727272727272727e-07, | |
| "logits/generated": -2.6749789714813232, | |
| "logits/real": -2.711810350418091, | |
| "logps/generated": -103.34049987792969, | |
| "logps/real": -90.66702270507812, | |
| "loss": 0.7685, | |
| "rewards/accuracies": 0.25, | |
| "rewards/generated": 0.17951178550720215, | |
| "rewards/margins": 0.16819272935390472, | |
| "rewards/real": 0.34770452976226807, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 62.39690097528854, | |
| "learning_rate": 3.818181818181818e-07, | |
| "logits/generated": -2.8889403343200684, | |
| "logits/real": -2.8932297229766846, | |
| "logps/generated": -112.69966125488281, | |
| "logps/real": -104.17449951171875, | |
| "loss": 0.7714, | |
| "rewards/accuracies": 0.2750000059604645, | |
| "rewards/generated": -0.4469524323940277, | |
| "rewards/margins": 0.310979425907135, | |
| "rewards/real": -0.1359730362892151, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 70.93062255354987, | |
| "learning_rate": 2.909090909090909e-07, | |
| "logits/generated": -3.0333216190338135, | |
| "logits/real": -3.0492300987243652, | |
| "logps/generated": -98.00860595703125, | |
| "logps/real": -83.57305908203125, | |
| "loss": 0.7209, | |
| "rewards/accuracies": 0.22499999403953552, | |
| "rewards/generated": 0.3839380145072937, | |
| "rewards/margins": 0.18280868232250214, | |
| "rewards/real": 0.566746711730957, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 55.19518628250898, | |
| "learning_rate": 2e-07, | |
| "logits/generated": -3.0203452110290527, | |
| "logits/real": -3.0042731761932373, | |
| "logps/generated": -129.9265594482422, | |
| "logps/real": -99.64141845703125, | |
| "loss": 0.7566, | |
| "rewards/accuracies": 0.23749999701976776, | |
| "rewards/generated": 0.8018711805343628, | |
| "rewards/margins": 0.49393320083618164, | |
| "rewards/real": 1.2958043813705444, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 21.78570359935915, | |
| "learning_rate": 1.0909090909090908e-07, | |
| "logits/generated": -2.9519670009613037, | |
| "logits/real": -2.9463257789611816, | |
| "logps/generated": -100.23809814453125, | |
| "logps/real": -82.25651550292969, | |
| "loss": 0.7277, | |
| "rewards/accuracies": 0.22499999403953552, | |
| "rewards/generated": 1.315015435218811, | |
| "rewards/margins": 0.4744125306606293, | |
| "rewards/real": 1.7894279956817627, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 55.38490015693705, | |
| "learning_rate": 1.818181818181818e-08, | |
| "logits/generated": -2.898479461669922, | |
| "logits/real": -2.902143955230713, | |
| "logps/generated": -86.09727478027344, | |
| "logps/real": -77.5208511352539, | |
| "loss": 0.7312, | |
| "rewards/accuracies": 0.13750000298023224, | |
| "rewards/generated": 1.5982306003570557, | |
| "rewards/margins": 0.5042404532432556, | |
| "rewards/real": 2.102470874786377, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "step": 62, | |
| "total_flos": 0.0, | |
| "train_loss": 0.7465078599991337, | |
| "train_runtime": 792.7009, | |
| "train_samples_per_second": 2.522, | |
| "train_steps_per_second": 0.078 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 62, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 200, | |
| "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 | |
| } | |