| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10488364470665355, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005244182235332678, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005244182235332678, | |
| "eval_loss": NaN, | |
| "eval_runtime": 166.1287, | |
| "eval_samples_per_second": 1.938, | |
| "eval_steps_per_second": 0.969, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010488364470665356, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 9719.7461, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.015732546705998034, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 4.5128, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02097672894133071, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 5.8465, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02622091117666339, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02622091117666339, | |
| "eval_loss": NaN, | |
| "eval_runtime": 22.0808, | |
| "eval_samples_per_second": 14.583, | |
| "eval_steps_per_second": 7.291, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03146509341199607, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.036709275647328746, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04195345788266142, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 15.1475, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0471976401179941, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05244182235332678, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05244182235332678, | |
| "eval_loss": NaN, | |
| "eval_runtime": 15.0673, | |
| "eval_samples_per_second": 21.371, | |
| "eval_steps_per_second": 10.685, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.057686004588659454, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 81.0289, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.06293018682399214, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 4195.0371, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06817436905932481, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.07341855129465749, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 35039.5117, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.07866273352999016, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07866273352999016, | |
| "eval_loss": NaN, | |
| "eval_runtime": 15.8266, | |
| "eval_samples_per_second": 20.346, | |
| "eval_steps_per_second": 10.173, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08390691576532285, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.08915109800065553, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.102, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0943952802359882, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09963946247132088, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.10488364470665355, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10488364470665355, | |
| "eval_loss": NaN, | |
| "eval_runtime": 14.775, | |
| "eval_samples_per_second": 21.794, | |
| "eval_steps_per_second": 10.897, | |
| "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": 1.4115215184494592e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |