| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.016664062906837473, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008332031453418737, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4743, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008332031453418737, | |
| "eval_loss": NaN, | |
| "eval_runtime": 130.2784, | |
| "eval_samples_per_second": 15.521, | |
| "eval_steps_per_second": 7.76, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016664062906837473, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.002499609436025621, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 1123.4865, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0033328125813674946, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 6910.874, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004166015726709368, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 49.9995, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004166015726709368, | |
| "eval_loss": NaN, | |
| "eval_runtime": 48.2159, | |
| "eval_samples_per_second": 41.936, | |
| "eval_steps_per_second": 20.968, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004999218872051242, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 54.1644, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005832422017393116, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 25.7448, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006665625162734989, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007498828308076863, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 148.3079, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008332031453418737, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 21.5046, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008332031453418737, | |
| "eval_loss": NaN, | |
| "eval_runtime": 48.8171, | |
| "eval_samples_per_second": 41.42, | |
| "eval_steps_per_second": 20.71, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00916523459876061, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 3591.5012, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009998437744102483, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.4636, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010831640889444358, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 778.5071, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011664844034786232, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.012498047180128105, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 208.5211, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012498047180128105, | |
| "eval_loss": NaN, | |
| "eval_runtime": 61.123, | |
| "eval_samples_per_second": 33.081, | |
| "eval_steps_per_second": 16.54, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.013331250325469978, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.1787, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.014164453470811852, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1265.0673, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.014997656616153727, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1601.2178, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0158308597614956, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 272.6218, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.016664062906837473, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 1.2589, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.016664062906837473, | |
| "eval_loss": NaN, | |
| "eval_runtime": 53.5787, | |
| "eval_samples_per_second": 37.739, | |
| "eval_steps_per_second": 18.869, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5422329007898624e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |