| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 1000, | |
| "global_step": 3528, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05668934240362812, | |
| "grad_norm": 0.3524502481986371, | |
| "learning_rate": 2.8328611898017e-06, | |
| "loss": 0.361, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11337868480725624, | |
| "grad_norm": 0.3997035804333986, | |
| "learning_rate": 5.6657223796034e-06, | |
| "loss": 0.3349, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17006802721088435, | |
| "grad_norm": 0.38751841177102114, | |
| "learning_rate": 8.4985835694051e-06, | |
| "loss": 0.332, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22675736961451248, | |
| "grad_norm": 0.44836449253905825, | |
| "learning_rate": 9.994594078481824e-06, | |
| "loss": 0.3352, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2834467120181406, | |
| "grad_norm": 0.4021994062977158, | |
| "learning_rate": 9.947201553489385e-06, | |
| "loss": 0.3324, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3401360544217687, | |
| "grad_norm": 0.3947990104528409, | |
| "learning_rate": 9.851412130607201e-06, | |
| "loss": 0.3352, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3968253968253968, | |
| "grad_norm": 0.4273504082938427, | |
| "learning_rate": 9.70816288726541e-06, | |
| "loss": 0.3357, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.45351473922902497, | |
| "grad_norm": 0.3916187294777505, | |
| "learning_rate": 9.518855185209205e-06, | |
| "loss": 0.3386, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5102040816326531, | |
| "grad_norm": 0.3417478841692644, | |
| "learning_rate": 9.285340961422978e-06, | |
| "loss": 0.3369, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5668934240362812, | |
| "grad_norm": 0.36668292635175, | |
| "learning_rate": 9.009904611220361e-06, | |
| "loss": 0.3386, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5668934240362812, | |
| "eval_loss": 0.3266855776309967, | |
| "eval_runtime": 713.495, | |
| "eval_samples_per_second": 4.395, | |
| "eval_steps_per_second": 0.549, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6235827664399093, | |
| "grad_norm": 0.49562486933370287, | |
| "learning_rate": 8.695240640732093e-06, | |
| "loss": 0.3351, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.6802721088435374, | |
| "grad_norm": 0.29570204380494114, | |
| "learning_rate": 8.344427307410323e-06, | |
| "loss": 0.3376, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7369614512471655, | |
| "grad_norm": 0.4078184129717466, | |
| "learning_rate": 7.960896506415892e-06, | |
| "loss": 0.335, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7936507936507936, | |
| "grad_norm": 0.3889595466840503, | |
| "learning_rate": 7.548400197480501e-06, | |
| "loss": 0.3347, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8503401360544217, | |
| "grad_norm": 0.3485999541903875, | |
| "learning_rate": 7.110973700679067e-06, | |
| "loss": 0.3346, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9070294784580499, | |
| "grad_norm": 0.3508468605704029, | |
| "learning_rate": 6.652896220178041e-06, | |
| "loss": 0.3377, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.963718820861678, | |
| "grad_norm": 0.3277279426137875, | |
| "learning_rate": 6.178648982143307e-06, | |
| "loss": 0.3284, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.0204081632653061, | |
| "grad_norm": 0.3946171462399535, | |
| "learning_rate": 5.692871396331168e-06, | |
| "loss": 0.3144, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.0770975056689343, | |
| "grad_norm": 0.36094113773422365, | |
| "learning_rate": 5.2003156702196275e-06, | |
| "loss": 0.276, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.1337868480725624, | |
| "grad_norm": 0.38888742960958356, | |
| "learning_rate": 4.705800319675462e-06, | |
| "loss": 0.2706, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1337868480725624, | |
| "eval_loss": 0.3267841041088104, | |
| "eval_runtime": 713.4781, | |
| "eval_samples_per_second": 4.395, | |
| "eval_steps_per_second": 0.549, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 0.3794229011992927, | |
| "learning_rate": 4.214163030947442e-06, | |
| "loss": 0.2785, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.2471655328798186, | |
| "grad_norm": 0.37402259299734353, | |
| "learning_rate": 3.730213335121796e-06, | |
| "loss": 0.2756, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.3038548752834467, | |
| "grad_norm": 0.3529299203177358, | |
| "learning_rate": 3.2586855580106876e-06, | |
| "loss": 0.2774, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.3605442176870748, | |
| "grad_norm": 0.3225567379873416, | |
| "learning_rate": 2.8041925057499253e-06, | |
| "loss": 0.2741, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.417233560090703, | |
| "grad_norm": 0.3853580675773505, | |
| "learning_rate": 2.371180339184999e-06, | |
| "loss": 0.2697, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.473922902494331, | |
| "grad_norm": 0.37229687058731975, | |
| "learning_rate": 1.963885078494927e-06, | |
| "loss": 0.2683, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.5306122448979593, | |
| "grad_norm": 0.38174797365201735, | |
| "learning_rate": 1.5862911635553614e-06, | |
| "loss": 0.2728, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.5873015873015874, | |
| "grad_norm": 0.43914992324605495, | |
| "learning_rate": 1.2420924754316915e-06, | |
| "loss": 0.2756, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.6439909297052155, | |
| "grad_norm": 0.3851735442216275, | |
| "learning_rate": 9.346562003165193e-07, | |
| "loss": 0.271, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.7006802721088436, | |
| "grad_norm": 0.463016547902216, | |
| "learning_rate": 6.669898894190413e-07, | |
| "loss": 0.2689, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7006802721088436, | |
| "eval_loss": 0.32290545105934143, | |
| "eval_runtime": 713.868, | |
| "eval_samples_per_second": 4.393, | |
| "eval_steps_per_second": 0.549, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7573696145124718, | |
| "grad_norm": 0.38528285238831594, | |
| "learning_rate": 4.417120370490169e-07, | |
| "loss": 0.2701, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.8140589569160999, | |
| "grad_norm": 0.4044768062755781, | |
| "learning_rate": 2.6102646472050463e-07, | |
| "loss": 0.2683, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.870748299319728, | |
| "grad_norm": 0.36435819425940585, | |
| "learning_rate": 1.2670076186756973e-07, | |
| "loss": 0.2732, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.927437641723356, | |
| "grad_norm": 0.38843363773407674, | |
| "learning_rate": 4.004899407954621e-08, | |
| "loss": 0.27, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.9841269841269842, | |
| "grad_norm": 0.4701515632923394, | |
| "learning_rate": 1.9188480157023014e-09, | |
| "loss": 0.2657, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 3528, | |
| "total_flos": 2304752043950080.0, | |
| "train_loss": 0.30443412133084946, | |
| "train_runtime": 78479.1521, | |
| "train_samples_per_second": 0.719, | |
| "train_steps_per_second": 0.045 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3528, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2304752043950080.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |