| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.820761762509335, | |
| "eval_steps": 1602, | |
| "global_step": 16020, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.982076176250933, | |
| "grad_norm": 1.8657052516937256, | |
| "learning_rate": 9.001248439450687e-06, | |
| "loss": 1.7286, | |
| "step": 1602 | |
| }, | |
| { | |
| "epoch": 5.982076176250933, | |
| "eval_accuracy": 0.06538245219347581, | |
| "eval_loss": 3.015066623687744, | |
| "eval_runtime": 37.6228, | |
| "eval_samples_per_second": 189.034, | |
| "eval_steps_per_second": 9.462, | |
| "step": 1602 | |
| }, | |
| { | |
| "epoch": 11.964152352501866, | |
| "grad_norm": 4.766932964324951, | |
| "learning_rate": 8.001248439450687e-06, | |
| "loss": 1.6207, | |
| "step": 3204 | |
| }, | |
| { | |
| "epoch": 11.964152352501866, | |
| "eval_accuracy": 0.06650731158605175, | |
| "eval_loss": 3.237614393234253, | |
| "eval_runtime": 37.2902, | |
| "eval_samples_per_second": 190.72, | |
| "eval_steps_per_second": 9.547, | |
| "step": 3204 | |
| }, | |
| { | |
| "epoch": 17.9462285287528, | |
| "grad_norm": 3.993530035018921, | |
| "learning_rate": 7.00187265917603e-06, | |
| "loss": 1.5399, | |
| "step": 4806 | |
| }, | |
| { | |
| "epoch": 17.9462285287528, | |
| "eval_accuracy": 0.06847581552305962, | |
| "eval_loss": 3.2385711669921875, | |
| "eval_runtime": 38.1712, | |
| "eval_samples_per_second": 186.319, | |
| "eval_steps_per_second": 9.326, | |
| "step": 4806 | |
| }, | |
| { | |
| "epoch": 23.928304705003733, | |
| "grad_norm": 1.5667791366577148, | |
| "learning_rate": 6.002496878901374e-06, | |
| "loss": 1.4981, | |
| "step": 6408 | |
| }, | |
| { | |
| "epoch": 23.928304705003733, | |
| "eval_accuracy": 0.0673158042744657, | |
| "eval_loss": 3.354517936706543, | |
| "eval_runtime": 37.7648, | |
| "eval_samples_per_second": 188.324, | |
| "eval_steps_per_second": 9.427, | |
| "step": 6408 | |
| }, | |
| { | |
| "epoch": 29.910380881254667, | |
| "grad_norm": 2.5913686752319336, | |
| "learning_rate": 5.003121098626717e-06, | |
| "loss": 1.4774, | |
| "step": 8010 | |
| }, | |
| { | |
| "epoch": 29.910380881254667, | |
| "eval_accuracy": 0.06766029246344207, | |
| "eval_loss": 3.340369939804077, | |
| "eval_runtime": 37.3411, | |
| "eval_samples_per_second": 190.461, | |
| "eval_steps_per_second": 9.534, | |
| "step": 8010 | |
| }, | |
| { | |
| "epoch": 35.8924570575056, | |
| "grad_norm": 1.792555332183838, | |
| "learning_rate": 4.003121098626717e-06, | |
| "loss": 1.4648, | |
| "step": 9612 | |
| }, | |
| { | |
| "epoch": 35.8924570575056, | |
| "eval_accuracy": 0.06699943757030372, | |
| "eval_loss": 3.4236271381378174, | |
| "eval_runtime": 37.2235, | |
| "eval_samples_per_second": 191.062, | |
| "eval_steps_per_second": 9.564, | |
| "step": 9612 | |
| }, | |
| { | |
| "epoch": 41.87453323375654, | |
| "grad_norm": 1.6825398206710815, | |
| "learning_rate": 3.003121098626717e-06, | |
| "loss": 1.4549, | |
| "step": 11214 | |
| }, | |
| { | |
| "epoch": 41.87453323375654, | |
| "eval_accuracy": 0.06644705126144947, | |
| "eval_loss": 3.4688560962677, | |
| "eval_runtime": 37.3243, | |
| "eval_samples_per_second": 190.546, | |
| "eval_steps_per_second": 9.538, | |
| "step": 11214 | |
| }, | |
| { | |
| "epoch": 47.856609410007465, | |
| "grad_norm": 1.0186405181884766, | |
| "learning_rate": 2.00374531835206e-06, | |
| "loss": 1.4528, | |
| "step": 12816 | |
| }, | |
| { | |
| "epoch": 47.856609410007465, | |
| "eval_accuracy": 0.06592730596175478, | |
| "eval_loss": 3.520540237426758, | |
| "eval_runtime": 37.6061, | |
| "eval_samples_per_second": 189.118, | |
| "eval_steps_per_second": 9.467, | |
| "step": 12816 | |
| }, | |
| { | |
| "epoch": 53.8386855862584, | |
| "grad_norm": 1.1254180669784546, | |
| "learning_rate": 1.0043695380774035e-06, | |
| "loss": 1.4538, | |
| "step": 14418 | |
| }, | |
| { | |
| "epoch": 53.8386855862584, | |
| "eval_accuracy": 0.06553868266466692, | |
| "eval_loss": 3.4702768325805664, | |
| "eval_runtime": 37.4351, | |
| "eval_samples_per_second": 189.982, | |
| "eval_steps_per_second": 9.51, | |
| "step": 14418 | |
| }, | |
| { | |
| "epoch": 59.820761762509335, | |
| "grad_norm": 0.6253169775009155, | |
| "learning_rate": 4.993757802746567e-09, | |
| "loss": 1.4519, | |
| "step": 16020 | |
| }, | |
| { | |
| "epoch": 59.820761762509335, | |
| "eval_accuracy": 0.06511529808773903, | |
| "eval_loss": 3.4956905841827393, | |
| "eval_runtime": 38.202, | |
| "eval_samples_per_second": 186.168, | |
| "eval_steps_per_second": 9.319, | |
| "step": 16020 | |
| }, | |
| { | |
| "epoch": 59.820761762509335, | |
| "step": 16020, | |
| "total_flos": 1.9001138663630822e+18, | |
| "train_loss": 1.5142938697233925, | |
| "train_runtime": 42053.4844, | |
| "train_samples_per_second": 76.397, | |
| "train_steps_per_second": 0.381 | |
| } | |
| ], | |
| "logging_steps": 1602, | |
| "max_steps": 16020, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "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": 1.9001138663630822e+18, | |
| "train_batch_size": 20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |