| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.011114198388441235, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005557099194220616, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 73.9887, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005557099194220616, | |
| "eval_loss": NaN, | |
| "eval_runtime": 562.7215, | |
| "eval_samples_per_second": 5.386, | |
| "eval_steps_per_second": 2.694, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011114198388441233, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0016671297582661851, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 2.8229, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0022228396776882466, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0027785495971103086, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0027785495971103086, | |
| "eval_loss": NaN, | |
| "eval_runtime": 273.9949, | |
| "eval_samples_per_second": 11.062, | |
| "eval_steps_per_second": 5.533, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0033342595165323703, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.003889969435954432, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.004445679355376493, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005001389274798555, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005557099194220617, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.005557099194220617, | |
| "eval_loss": NaN, | |
| "eval_runtime": 273.6936, | |
| "eval_samples_per_second": 11.074, | |
| "eval_steps_per_second": 5.539, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006112809113642679, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0066685190330647405, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007224228952486802, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.007779938871908864, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.3996, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.008335648791330925, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008335648791330925, | |
| "eval_loss": NaN, | |
| "eval_runtime": 276.6536, | |
| "eval_samples_per_second": 10.956, | |
| "eval_steps_per_second": 5.48, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008891358710752986, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 22.0925, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.009447068630175049, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 19.9371, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01000277854959711, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.010558488469019172, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 36.846, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.011114198388441235, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011114198388441235, | |
| "eval_loss": NaN, | |
| "eval_runtime": 304.0478, | |
| "eval_samples_per_second": 9.969, | |
| "eval_steps_per_second": 4.986, | |
| "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": 2.4960350734516224e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |