| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 4035, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 13.118809700012207, | |
| "learning_rate": 9.335811648079307e-06, | |
| "loss": 0.8716, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6197183132171631, | |
| "eval_loss": 0.9822000861167908, | |
| "eval_runtime": 32.2509, | |
| "eval_samples_per_second": 2.201, | |
| "eval_steps_per_second": 2.201, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 10.717401504516602, | |
| "learning_rate": 8.66914498141264e-06, | |
| "loss": 0.8143, | |
| "step": 538 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5352112650871277, | |
| "eval_loss": 1.232447624206543, | |
| "eval_runtime": 32.3951, | |
| "eval_samples_per_second": 2.192, | |
| "eval_steps_per_second": 2.192, | |
| "step": 538 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 23.701391220092773, | |
| "learning_rate": 8.004956629491945e-06, | |
| "loss": 0.7584, | |
| "step": 807 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6478873491287231, | |
| "eval_loss": 1.0226496458053589, | |
| "eval_runtime": 32.3, | |
| "eval_samples_per_second": 2.198, | |
| "eval_steps_per_second": 2.198, | |
| "step": 807 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 33.54118728637695, | |
| "learning_rate": 7.340768277571252e-06, | |
| "loss": 0.6715, | |
| "step": 1076 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6619718074798584, | |
| "eval_loss": 0.9550462961196899, | |
| "eval_runtime": 32.4715, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 2.187, | |
| "step": 1076 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 30.0217227935791, | |
| "learning_rate": 6.674101610904585e-06, | |
| "loss": 0.6471, | |
| "step": 1345 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6760563254356384, | |
| "eval_loss": 1.1272403001785278, | |
| "eval_runtime": 32.2183, | |
| "eval_samples_per_second": 2.204, | |
| "eval_steps_per_second": 2.204, | |
| "step": 1345 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.36277323961257935, | |
| "learning_rate": 6.009913258983892e-06, | |
| "loss": 0.5759, | |
| "step": 1614 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6760563254356384, | |
| "eval_loss": 1.2193043231964111, | |
| "eval_runtime": 32.3011, | |
| "eval_samples_per_second": 2.198, | |
| "eval_steps_per_second": 2.198, | |
| "step": 1614 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 17.32307243347168, | |
| "learning_rate": 5.343246592317225e-06, | |
| "loss": 0.4963, | |
| "step": 1883 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7183098793029785, | |
| "eval_loss": 1.221394658088684, | |
| "eval_runtime": 32.1815, | |
| "eval_samples_per_second": 2.206, | |
| "eval_steps_per_second": 2.206, | |
| "step": 1883 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.09576527029275894, | |
| "learning_rate": 4.679058240396531e-06, | |
| "loss": 0.4053, | |
| "step": 2152 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7464788556098938, | |
| "eval_loss": 1.3082976341247559, | |
| "eval_runtime": 32.7134, | |
| "eval_samples_per_second": 2.17, | |
| "eval_steps_per_second": 2.17, | |
| "step": 2152 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 73.52884674072266, | |
| "learning_rate": 4.012391573729864e-06, | |
| "loss": 0.3344, | |
| "step": 2421 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6619718074798584, | |
| "eval_loss": 1.6390645503997803, | |
| "eval_runtime": 32.8911, | |
| "eval_samples_per_second": 2.159, | |
| "eval_steps_per_second": 2.159, | |
| "step": 2421 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 242.48890686035156, | |
| "learning_rate": 3.34820322180917e-06, | |
| "loss": 0.3216, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6478873491287231, | |
| "eval_loss": 1.722383737564087, | |
| "eval_runtime": 32.4813, | |
| "eval_samples_per_second": 2.186, | |
| "eval_steps_per_second": 2.186, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 376.0904541015625, | |
| "learning_rate": 2.6815365551425034e-06, | |
| "loss": 0.2248, | |
| "step": 2959 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6760563254356384, | |
| "eval_loss": 1.7972948551177979, | |
| "eval_runtime": 32.2962, | |
| "eval_samples_per_second": 2.198, | |
| "eval_steps_per_second": 2.198, | |
| "step": 2959 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 38.31782913208008, | |
| "learning_rate": 2.0148698884758364e-06, | |
| "loss": 0.1982, | |
| "step": 3228 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6478873491287231, | |
| "eval_loss": 2.02411150932312, | |
| "eval_runtime": 32.4081, | |
| "eval_samples_per_second": 2.191, | |
| "eval_steps_per_second": 2.191, | |
| "step": 3228 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.5905938744544983, | |
| "learning_rate": 1.3482032218091697e-06, | |
| "loss": 0.1362, | |
| "step": 3497 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6478873491287231, | |
| "eval_loss": 1.9932571649551392, | |
| "eval_runtime": 32.4096, | |
| "eval_samples_per_second": 2.191, | |
| "eval_steps_per_second": 2.191, | |
| "step": 3497 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 2.5972626209259033, | |
| "learning_rate": 6.815365551425032e-07, | |
| "loss": 0.0879, | |
| "step": 3766 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6478873491287231, | |
| "eval_loss": 2.086475372314453, | |
| "eval_runtime": 32.8946, | |
| "eval_samples_per_second": 2.158, | |
| "eval_steps_per_second": 2.158, | |
| "step": 3766 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.036518827080726624, | |
| "learning_rate": 1.4869888475836432e-08, | |
| "loss": 0.0712, | |
| "step": 4035 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.6338028311729431, | |
| "eval_loss": 2.1690945625305176, | |
| "eval_runtime": 32.3643, | |
| "eval_samples_per_second": 2.194, | |
| "eval_steps_per_second": 2.194, | |
| "step": 4035 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 4035, | |
| "total_flos": 7.1272809219168e+18, | |
| "train_loss": 0.44098070326643213, | |
| "train_runtime": 9434.6926, | |
| "train_samples_per_second": 1.283, | |
| "train_steps_per_second": 0.428 | |
| } | |
| ], | |
| "logging_steps": 8, | |
| "max_steps": 4035, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.1272809219168e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |