| { | |
| "best_metric": 1.3715944290161133, | |
| "best_model_checkpoint": "flan-t5-base-samsum/checkpoint-3684", | |
| "epoch": 5.0, | |
| "global_step": 9210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.728555917480999e-05, | |
| "loss": 1.4811, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.457111834961998e-05, | |
| "loss": 1.4545, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.185667752442997e-05, | |
| "loss": 1.4379, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 17.11965811965812, | |
| "eval_loss": 1.3805454969406128, | |
| "eval_rouge1": 47.1438, | |
| "eval_rouge2": 23.6153, | |
| "eval_rougeL": 39.6699, | |
| "eval_rougeLsum": 43.5505, | |
| "eval_runtime": 126.1042, | |
| "eval_samples_per_second": 6.495, | |
| "eval_steps_per_second": 0.817, | |
| "step": 1842 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3.914223669923996e-05, | |
| "loss": 1.4094, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 3.642779587404995e-05, | |
| "loss": 1.3318, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.3713355048859935e-05, | |
| "loss": 1.3396, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.099891422366993e-05, | |
| "loss": 1.3559, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 17.39072039072039, | |
| "eval_loss": 1.3715944290161133, | |
| "eval_rouge1": 47.2663, | |
| "eval_rouge2": 23.5327, | |
| "eval_rougeL": 39.6491, | |
| "eval_rougeLsum": 43.3169, | |
| "eval_runtime": 125.9833, | |
| "eval_samples_per_second": 6.501, | |
| "eval_steps_per_second": 0.818, | |
| "step": 3684 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 2.8284473398479917e-05, | |
| "loss": 1.3123, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 2.5570032573289905e-05, | |
| "loss": 1.2724, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 2.2855591748099893e-05, | |
| "loss": 1.2815, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 2.0141150922909884e-05, | |
| "loss": 1.2783, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 17.183150183150182, | |
| "eval_loss": 1.3721352815628052, | |
| "eval_rouge1": 47.4896, | |
| "eval_rouge2": 23.7684, | |
| "eval_rougeL": 39.7733, | |
| "eval_rougeLsum": 43.4494, | |
| "eval_runtime": 125.6835, | |
| "eval_samples_per_second": 6.516, | |
| "eval_steps_per_second": 0.82, | |
| "step": 5526 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 1.742671009771987e-05, | |
| "loss": 1.2228, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 1.471226927252986e-05, | |
| "loss": 1.243, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.1997828447339848e-05, | |
| "loss": 1.2378, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 17.305250305250304, | |
| "eval_loss": 1.3757330179214478, | |
| "eval_rouge1": 47.9122, | |
| "eval_rouge2": 24.0531, | |
| "eval_rougeL": 40.2225, | |
| "eval_rougeLsum": 43.996, | |
| "eval_runtime": 125.846, | |
| "eval_samples_per_second": 6.508, | |
| "eval_steps_per_second": 0.818, | |
| "step": 7368 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 9.283387622149838e-06, | |
| "loss": 1.2306, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 6.568946796959827e-06, | |
| "loss": 1.2106, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 3.854505971769816e-06, | |
| "loss": 1.1991, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.1400651465798045e-06, | |
| "loss": 1.1983, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 17.304029304029303, | |
| "eval_loss": 1.3750557899475098, | |
| "eval_rouge1": 47.8507, | |
| "eval_rouge2": 24.0061, | |
| "eval_rougeL": 40.231, | |
| "eval_rougeLsum": 43.8698, | |
| "eval_runtime": 125.3005, | |
| "eval_samples_per_second": 6.536, | |
| "eval_steps_per_second": 0.822, | |
| "step": 9210 | |
| } | |
| ], | |
| "max_steps": 9210, | |
| "num_train_epochs": 5, | |
| "total_flos": 5.043922658131968e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |