| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7497374868743437, | |
| "eval_steps": 1142, | |
| "global_step": 4284, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08750437521876094, | |
| "grad_norm": 8.797028541564941, | |
| "learning_rate": 1.824991249562478e-05, | |
| "loss": 3.9704, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17500875043752187, | |
| "grad_norm": 7.093753337860107, | |
| "learning_rate": 1.6499824991249566e-05, | |
| "loss": 3.6946, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19985999299965, | |
| "eval_gen_len": 36.008, | |
| "eval_loss": 3.3878133296966553, | |
| "eval_rouge1": 0.1544, | |
| "eval_rouge2": 0.0339, | |
| "eval_rougeL": 0.1176, | |
| "eval_rougeLsum": 0.1369, | |
| "eval_runtime": 204.0944, | |
| "eval_samples_per_second": 2.45, | |
| "eval_steps_per_second": 0.157, | |
| "step": 1142 | |
| }, | |
| { | |
| "epoch": 0.2625131256562828, | |
| "grad_norm": 9.073663711547852, | |
| "learning_rate": 1.4749737486874345e-05, | |
| "loss": 3.6103, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.35001750087504374, | |
| "grad_norm": 6.332721710205078, | |
| "learning_rate": 1.2999649982499125e-05, | |
| "loss": 3.5426, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3997199859993, | |
| "eval_gen_len": 35.828, | |
| "eval_loss": 3.291989326477051, | |
| "eval_rouge1": 0.1731, | |
| "eval_rouge2": 0.0402, | |
| "eval_rougeL": 0.1319, | |
| "eval_rougeLsum": 0.1527, | |
| "eval_runtime": 208.9687, | |
| "eval_samples_per_second": 2.393, | |
| "eval_steps_per_second": 0.153, | |
| "step": 2284 | |
| }, | |
| { | |
| "epoch": 0.43752187609380466, | |
| "grad_norm": 6.205784320831299, | |
| "learning_rate": 1.1249562478123908e-05, | |
| "loss": 3.4922, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5250262513125656, | |
| "grad_norm": 6.974767208099365, | |
| "learning_rate": 9.499474973748688e-06, | |
| "loss": 3.4693, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.59957997899895, | |
| "eval_gen_len": 40.762, | |
| "eval_loss": 3.243711233139038, | |
| "eval_rouge1": 0.1798, | |
| "eval_rouge2": 0.0435, | |
| "eval_rougeL": 0.1346, | |
| "eval_rougeLsum": 0.1589, | |
| "eval_runtime": 213.904, | |
| "eval_samples_per_second": 2.337, | |
| "eval_steps_per_second": 0.15, | |
| "step": 3426 | |
| }, | |
| { | |
| "epoch": 0.6125306265313266, | |
| "grad_norm": 6.394979953765869, | |
| "learning_rate": 7.749387469373469e-06, | |
| "loss": 3.4334, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7000350017500875, | |
| "grad_norm": 5.80218505859375, | |
| "learning_rate": 5.999299964998251e-06, | |
| "loss": 3.4327, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5714, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 714, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.501363131899904e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |