| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7321835340058575, | |
| "global_step": 13500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.981921394222078e-05, | |
| "loss": 0.1383, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.963842788444155e-05, | |
| "loss": 0.1388, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.945764182666233e-05, | |
| "loss": 0.1394, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.92768557688831e-05, | |
| "loss": 0.1395, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.909606971110389e-05, | |
| "loss": 0.1401, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.891528365332466e-05, | |
| "loss": 0.1402, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.873449759554544e-05, | |
| "loss": 0.1404, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.855371153776621e-05, | |
| "loss": 0.1404, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.837292547998698e-05, | |
| "loss": 0.1404, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_bleu": 58.9732, | |
| "eval_gen_len": 127.0, | |
| "eval_loss": 0.16096045076847076, | |
| "eval_max_gen_len": 127.0, | |
| "eval_runtime": 5905.2845, | |
| "eval_samples_per_second": 2.54, | |
| "eval_steps_per_second": 0.053, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.819213942220776e-05, | |
| "loss": 0.1407, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.801135336442853e-05, | |
| "loss": 0.1408, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.78305673066493e-05, | |
| "loss": 0.1414, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 9.764978124887008e-05, | |
| "loss": 0.1412, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.746899519109087e-05, | |
| "loss": 0.1412, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 9.728820913331164e-05, | |
| "loss": 0.1414, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9.710742307553242e-05, | |
| "loss": 0.1414, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 9.69266370177532e-05, | |
| "loss": 0.1417, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 9.674585095997398e-05, | |
| "loss": 0.1412, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_bleu": 58.854, | |
| "eval_gen_len": 126.9273, | |
| "eval_loss": 0.1599324494600296, | |
| "eval_max_gen_len": 127.0, | |
| "eval_runtime": 5913.6366, | |
| "eval_samples_per_second": 2.537, | |
| "eval_steps_per_second": 0.053, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.656506490219475e-05, | |
| "loss": 0.1414, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.638427884441553e-05, | |
| "loss": 0.1414, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 9.62034927866363e-05, | |
| "loss": 0.1416, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 9.602270672885707e-05, | |
| "loss": 0.1417, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 9.584192067107786e-05, | |
| "loss": 0.1413, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 9.566113461329863e-05, | |
| "loss": 0.1414, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 9.548034855551941e-05, | |
| "loss": 0.1419, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 9.529956249774018e-05, | |
| "loss": 0.1416, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 9.511877643996096e-05, | |
| "loss": 0.1415, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_bleu": 58.7355, | |
| "eval_gen_len": 127.0, | |
| "eval_loss": 0.15812213718891144, | |
| "eval_max_gen_len": 127.0, | |
| "eval_runtime": 5912.1982, | |
| "eval_samples_per_second": 2.537, | |
| "eval_steps_per_second": 0.053, | |
| "step": 13500 | |
| } | |
| ], | |
| "max_steps": 276570, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.402764131303424e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |