| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.033631108775617445, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016815554387808723, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016815554387808723, | |
| "eval_loss": NaN, | |
| "eval_runtime": 155.6833, | |
| "eval_samples_per_second": 6.436, | |
| "eval_steps_per_second": 3.218, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0033631108775617445, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 4.3372, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005044666316342617, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006726221755123489, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.008407777193904361, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008407777193904361, | |
| "eval_loss": NaN, | |
| "eval_runtime": 71.3564, | |
| "eval_samples_per_second": 14.042, | |
| "eval_steps_per_second": 7.021, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010089332632685234, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011770888071466106, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.013452443510246978, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01513399894902785, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.016815554387808723, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016815554387808723, | |
| "eval_loss": NaN, | |
| "eval_runtime": 69.7929, | |
| "eval_samples_per_second": 14.357, | |
| "eval_steps_per_second": 7.178, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.018497109826589597, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.020178665265370467, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02186022070415134, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02354177614293221, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.025223331581713086, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 18594.6387, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.025223331581713086, | |
| "eval_loss": NaN, | |
| "eval_runtime": 71.5026, | |
| "eval_samples_per_second": 14.013, | |
| "eval_steps_per_second": 7.007, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.026904887020493956, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 33915.1875, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02858644245927483, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0302679978980557, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03194955333683657, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.033631108775617445, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.033631108775617445, | |
| "eval_loss": NaN, | |
| "eval_runtime": 71.5452, | |
| "eval_samples_per_second": 14.005, | |
| "eval_steps_per_second": 7.003, | |
| "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": 1.4415708989423616e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |