| { | |
| "best_metric": 0.6567384004592896, | |
| "best_model_checkpoint": "distilbert_lda_50_v1_mnli/checkpoint-4602", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 12272, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.3023266792297363, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.8292, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6905756495160469, | |
| "eval_loss": 0.7242587208747864, | |
| "eval_runtime": 4.1868, | |
| "eval_samples_per_second": 2344.298, | |
| "eval_steps_per_second": 9.315, | |
| "step": 1534 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.6002289056777954, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6692, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7184921039225675, | |
| "eval_loss": 0.6628246307373047, | |
| "eval_runtime": 4.2185, | |
| "eval_samples_per_second": 2326.655, | |
| "eval_steps_per_second": 9.245, | |
| "step": 3068 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.015767812728882, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5756, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7335710646968925, | |
| "eval_loss": 0.6567384004592896, | |
| "eval_runtime": 4.2186, | |
| "eval_samples_per_second": 2326.613, | |
| "eval_steps_per_second": 9.245, | |
| "step": 4602 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.5160274505615234, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.4939, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7320427916454406, | |
| "eval_loss": 0.6775555610656738, | |
| "eval_runtime": 4.1905, | |
| "eval_samples_per_second": 2342.201, | |
| "eval_steps_per_second": 9.307, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.3854494094848633, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.414, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7373408048904737, | |
| "eval_loss": 0.6913235187530518, | |
| "eval_runtime": 4.185, | |
| "eval_samples_per_second": 2345.269, | |
| "eval_steps_per_second": 9.319, | |
| "step": 7670 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.9510507583618164, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3422, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7314314824248599, | |
| "eval_loss": 0.7808761596679688, | |
| "eval_runtime": 4.2144, | |
| "eval_samples_per_second": 2328.947, | |
| "eval_steps_per_second": 9.254, | |
| "step": 9204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.713484048843384, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2807, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7349974528782476, | |
| "eval_loss": 0.8293947577476501, | |
| "eval_runtime": 4.1995, | |
| "eval_samples_per_second": 2337.177, | |
| "eval_steps_per_second": 9.287, | |
| "step": 10738 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.71089506149292, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2292, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7279673968415691, | |
| "eval_loss": 0.9724929928779602, | |
| "eval_runtime": 4.2505, | |
| "eval_samples_per_second": 2309.13, | |
| "eval_steps_per_second": 9.175, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 12272, | |
| "total_flos": 2.080845603745874e+17, | |
| "train_loss": 0.479245346248383, | |
| "train_runtime": 2964.6777, | |
| "train_samples_per_second": 6623.013, | |
| "train_steps_per_second": 25.871 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 76700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.080845603745874e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |