| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2882882882882883, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014414414414414415, | |
| "grad_norm": 2.5799503326416016, | |
| "learning_rate": 1e-05, | |
| "loss": 23.6318, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014414414414414415, | |
| "eval_loss": NaN, | |
| "eval_runtime": 128.8315, | |
| "eval_samples_per_second": 0.908, | |
| "eval_steps_per_second": 0.458, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02882882882882883, | |
| "grad_norm": 3.7943718433380127, | |
| "learning_rate": 2e-05, | |
| "loss": 26.805, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.043243243243243246, | |
| "grad_norm": 4.626277923583984, | |
| "learning_rate": 3e-05, | |
| "loss": 27.3763, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05765765765765766, | |
| "grad_norm": 3.004248857498169, | |
| "learning_rate": 4e-05, | |
| "loss": 23.8077, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07207207207207207, | |
| "grad_norm": 2.720515727996826, | |
| "learning_rate": 5e-05, | |
| "loss": 24.3285, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07207207207207207, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.3853, | |
| "eval_samples_per_second": 15.842, | |
| "eval_steps_per_second": 7.989, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08648648648648649, | |
| "grad_norm": 4.533637046813965, | |
| "learning_rate": 6e-05, | |
| "loss": 26.1821, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1009009009009009, | |
| "grad_norm": 4.709778785705566, | |
| "learning_rate": 7e-05, | |
| "loss": 26.2916, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11531531531531532, | |
| "grad_norm": 5.645783424377441, | |
| "learning_rate": 8e-05, | |
| "loss": 28.9214, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.12972972972972974, | |
| "grad_norm": 5.6830854415893555, | |
| "learning_rate": 9e-05, | |
| "loss": 27.7614, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14414414414414414, | |
| "grad_norm": 4.149396896362305, | |
| "learning_rate": 0.0001, | |
| "loss": 22.8343, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14414414414414414, | |
| "eval_loss": NaN, | |
| "eval_runtime": 11.5843, | |
| "eval_samples_per_second": 10.1, | |
| "eval_steps_per_second": 5.093, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15855855855855855, | |
| "grad_norm": 5.578166961669922, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 25.1102, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.17297297297297298, | |
| "grad_norm": 6.154359817504883, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 23.233, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1873873873873874, | |
| "grad_norm": 6.816771507263184, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 22.2927, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2018018018018018, | |
| "grad_norm": 7.438177585601807, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 22.9498, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.21621621621621623, | |
| "grad_norm": 4.74092960357666, | |
| "learning_rate": 5e-05, | |
| "loss": 20.1686, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.21621621621621623, | |
| "eval_loss": NaN, | |
| "eval_runtime": 11.296, | |
| "eval_samples_per_second": 10.358, | |
| "eval_steps_per_second": 5.223, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.23063063063063063, | |
| "grad_norm": 5.9180169105529785, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 23.0958, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.24504504504504504, | |
| "grad_norm": 5.075774669647217, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 20.6796, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2594594594594595, | |
| "grad_norm": 8.858704566955566, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 21.7613, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.27387387387387385, | |
| "grad_norm": 5.454452991485596, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 20.0674, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2882882882882883, | |
| "grad_norm": 7.088149547576904, | |
| "learning_rate": 0.0, | |
| "loss": 20.7839, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2882882882882883, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.0773, | |
| "eval_samples_per_second": 16.532, | |
| "eval_steps_per_second": 8.337, | |
| "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": 2.53297944428544e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |