| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "global_step": 11120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.8501199040767384e-05, | |
| "loss": 2.0957, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.700239808153478e-05, | |
| "loss": 2.022, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.550359712230216e-05, | |
| "loss": 1.9842, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.400479616306955e-05, | |
| "loss": 1.9624, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.2505995203836936e-05, | |
| "loss": 1.9237, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 25.317, | |
| "eval_gen_len": 11.9956, | |
| "eval_loss": 1.8112746477127075, | |
| "eval_runtime": 164.477, | |
| "eval_samples_per_second": 12.518, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.100719424460432e-05, | |
| "loss": 1.7808, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.9508393285371705e-05, | |
| "loss": 1.6002, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 3.8009592326139094e-05, | |
| "loss": 1.6191, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.6510791366906475e-05, | |
| "loss": 1.6215, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 3.501199040767386e-05, | |
| "loss": 1.6134, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 3.351318944844125e-05, | |
| "loss": 1.6013, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 25.523, | |
| "eval_gen_len": 12.0345, | |
| "eval_loss": 1.8125885725021362, | |
| "eval_runtime": 182.336, | |
| "eval_samples_per_second": 11.292, | |
| "step": 5560 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.201438848920863e-05, | |
| "loss": 1.3405, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.051558752997602e-05, | |
| "loss": 1.3232, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 2.9016786570743405e-05, | |
| "loss": 1.3278, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 2.7517985611510793e-05, | |
| "loss": 1.3295, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.6019184652278178e-05, | |
| "loss": 1.335, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 25.7471, | |
| "eval_gen_len": 11.8786, | |
| "eval_loss": 1.90364408493042, | |
| "eval_runtime": 179.3491, | |
| "eval_samples_per_second": 11.48, | |
| "step": 8340 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 2.4520383693045566e-05, | |
| "loss": 1.2511, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 2.302158273381295e-05, | |
| "loss": 1.0838, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 2.1522781774580335e-05, | |
| "loss": 1.0898, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 2.0023980815347724e-05, | |
| "loss": 1.0904, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.8525179856115108e-05, | |
| "loss": 1.0924, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.7026378896882493e-05, | |
| "loss": 1.0952, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 24.7279, | |
| "eval_gen_len": 12.0661, | |
| "eval_loss": 2.009955406188965, | |
| "eval_runtime": 180.8445, | |
| "eval_samples_per_second": 11.385, | |
| "step": 11120 | |
| } | |
| ], | |
| "max_steps": 16680, | |
| "num_train_epochs": 6, | |
| "total_flos": 1.2244523192274125e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |