| { | |
| "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": 187.93083613602994, | |
| "learning_rate": 7.142857142857142e-08, | |
| "logits/generated": -2.834711790084839, | |
| "logits/real": -3.0148138999938965, | |
| "logps/generated": -199.41073608398438, | |
| "logps/real": -229.6434326171875, | |
| "loss": 0.9151, | |
| "rewards/accuracies": 0.0, | |
| "rewards/generated": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/real": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 137.95931851523642, | |
| "learning_rate": 4.727272727272727e-07, | |
| "logits/generated": -2.88613224029541, | |
| "logits/real": -2.9299046993255615, | |
| "logps/generated": -221.88681030273438, | |
| "logps/real": -211.3094940185547, | |
| "loss": 0.872, | |
| "rewards/accuracies": 0.5416666865348816, | |
| "rewards/generated": 0.013987368904054165, | |
| "rewards/margins": 0.1755019873380661, | |
| "rewards/real": 0.18948933482170105, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 111.51574233912503, | |
| "learning_rate": 3.818181818181818e-07, | |
| "logits/generated": -2.764463186264038, | |
| "logits/real": -2.8043417930603027, | |
| "logps/generated": -241.7828369140625, | |
| "logps/real": -226.93673706054688, | |
| "loss": 0.8295, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/generated": -1.7762424945831299, | |
| "rewards/margins": 0.9314602017402649, | |
| "rewards/real": -0.8447820544242859, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 85.31778142133322, | |
| "learning_rate": 2.909090909090909e-07, | |
| "logits/generated": -2.688220262527466, | |
| "logits/real": -2.6674160957336426, | |
| "logps/generated": -256.388427734375, | |
| "logps/real": -225.5462188720703, | |
| "loss": 0.6307, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/generated": -2.2042856216430664, | |
| "rewards/margins": 1.7098881006240845, | |
| "rewards/real": -0.4943977892398834, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 99.66362503152259, | |
| "learning_rate": 2e-07, | |
| "logits/generated": -2.6126770973205566, | |
| "logits/real": -2.4645614624023438, | |
| "logps/generated": -249.4606475830078, | |
| "logps/real": -212.3032684326172, | |
| "loss": 0.5484, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/generated": -2.520421266555786, | |
| "rewards/margins": 2.207958459854126, | |
| "rewards/real": -0.3124631941318512, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 117.4078873382634, | |
| "learning_rate": 1.0909090909090908e-07, | |
| "logits/generated": -2.496239185333252, | |
| "logits/real": -2.617043972015381, | |
| "logps/generated": -253.89047241210938, | |
| "logps/real": -228.7570037841797, | |
| "loss": 0.51, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/generated": -3.078453302383423, | |
| "rewards/margins": 2.4696271419525146, | |
| "rewards/real": -0.6088260412216187, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 120.65322478839221, | |
| "learning_rate": 1.818181818181818e-08, | |
| "logits/generated": -2.538569688796997, | |
| "logits/real": -2.4903368949890137, | |
| "logps/generated": -258.3146057128906, | |
| "logps/real": -220.8245849609375, | |
| "loss": 0.545, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/generated": -3.1993417739868164, | |
| "rewards/margins": 2.3859920501708984, | |
| "rewards/real": -0.8133499026298523, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "step": 62, | |
| "total_flos": 0.0, | |
| "train_loss": 0.6561146609244808, | |
| "train_runtime": 788.6021, | |
| "train_samples_per_second": 2.535, | |
| "train_steps_per_second": 0.079 | |
| } | |
| ], | |
| "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 | |
| } | |