| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 18160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.9003303964757713e-05, | |
| "loss": 1.3504, | |
| "step": 908 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.9396927356719971, | |
| "eval_runtime": 40.3444, | |
| "eval_samples_per_second": 80.011, | |
| "eval_steps_per_second": 2.503, | |
| "step": 908 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.800330396475771e-05, | |
| "loss": 0.952, | |
| "step": 1816 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7900779247283936, | |
| "eval_runtime": 40.474, | |
| "eval_samples_per_second": 79.755, | |
| "eval_steps_per_second": 2.495, | |
| "step": 1816 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.700330396475771e-05, | |
| "loss": 0.8225, | |
| "step": 2724 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7042744755744934, | |
| "eval_runtime": 40.4219, | |
| "eval_samples_per_second": 79.858, | |
| "eval_steps_per_second": 2.499, | |
| "step": 2724 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.600330396475771e-05, | |
| "loss": 0.7475, | |
| "step": 3632 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6635274291038513, | |
| "eval_runtime": 40.4218, | |
| "eval_samples_per_second": 79.858, | |
| "eval_steps_per_second": 2.499, | |
| "step": 3632 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.5003303964757709e-05, | |
| "loss": 0.6945, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6002705693244934, | |
| "eval_runtime": 40.5407, | |
| "eval_samples_per_second": 79.624, | |
| "eval_steps_per_second": 2.491, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.4003303964757711e-05, | |
| "loss": 0.6558, | |
| "step": 5448 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.5814478993415833, | |
| "eval_runtime": 40.4802, | |
| "eval_samples_per_second": 79.743, | |
| "eval_steps_per_second": 2.495, | |
| "step": 5448 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.3003303964757709e-05, | |
| "loss": 0.6179, | |
| "step": 6356 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.530373215675354, | |
| "eval_runtime": 40.4721, | |
| "eval_samples_per_second": 79.759, | |
| "eval_steps_per_second": 2.496, | |
| "step": 6356 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.2004405286343613e-05, | |
| "loss": 0.5939, | |
| "step": 7264 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.5254124402999878, | |
| "eval_runtime": 40.4437, | |
| "eval_samples_per_second": 79.815, | |
| "eval_steps_per_second": 2.497, | |
| "step": 7264 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1.1004405286343614e-05, | |
| "loss": 0.5724, | |
| "step": 8172 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.5123583078384399, | |
| "eval_runtime": 40.4461, | |
| "eval_samples_per_second": 79.81, | |
| "eval_steps_per_second": 2.497, | |
| "step": 8172 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.0005506607929516e-05, | |
| "loss": 0.5518, | |
| "step": 9080 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.5051878094673157, | |
| "eval_runtime": 40.5153, | |
| "eval_samples_per_second": 79.674, | |
| "eval_steps_per_second": 2.493, | |
| "step": 9080 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 9.005506607929516e-06, | |
| "loss": 0.5374, | |
| "step": 9988 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.48619189858436584, | |
| "eval_runtime": 40.4854, | |
| "eval_samples_per_second": 79.732, | |
| "eval_steps_per_second": 2.495, | |
| "step": 9988 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 8.006607929515418e-06, | |
| "loss": 0.5257, | |
| "step": 10896 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.4822216033935547, | |
| "eval_runtime": 40.4555, | |
| "eval_samples_per_second": 79.791, | |
| "eval_steps_per_second": 2.497, | |
| "step": 10896 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 7.006607929515419e-06, | |
| "loss": 0.5109, | |
| "step": 11804 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.4707971215248108, | |
| "eval_runtime": 40.4351, | |
| "eval_samples_per_second": 79.832, | |
| "eval_steps_per_second": 2.498, | |
| "step": 11804 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 6.006607929515419e-06, | |
| "loss": 0.4991, | |
| "step": 12712 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.4686879813671112, | |
| "eval_runtime": 40.4561, | |
| "eval_samples_per_second": 79.79, | |
| "eval_steps_per_second": 2.497, | |
| "step": 12712 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 5.0066079295154195e-06, | |
| "loss": 0.4966, | |
| "step": 13620 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.4615265130996704, | |
| "eval_runtime": 40.492, | |
| "eval_samples_per_second": 79.719, | |
| "eval_steps_per_second": 2.494, | |
| "step": 13620 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 4.0077092511013225e-06, | |
| "loss": 0.4877, | |
| "step": 14528 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.45428407192230225, | |
| "eval_runtime": 40.5187, | |
| "eval_samples_per_second": 79.667, | |
| "eval_steps_per_second": 2.493, | |
| "step": 14528 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 3.007709251101322e-06, | |
| "loss": 0.4837, | |
| "step": 15436 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.4501075744628906, | |
| "eval_runtime": 40.541, | |
| "eval_samples_per_second": 79.623, | |
| "eval_steps_per_second": 2.491, | |
| "step": 15436 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2.0077092511013217e-06, | |
| "loss": 0.4788, | |
| "step": 16344 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.44279614090919495, | |
| "eval_runtime": 40.6464, | |
| "eval_samples_per_second": 79.417, | |
| "eval_steps_per_second": 2.485, | |
| "step": 16344 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.0088105726872247e-06, | |
| "loss": 0.4746, | |
| "step": 17252 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.4437817931175232, | |
| "eval_runtime": 40.5435, | |
| "eval_samples_per_second": 79.618, | |
| "eval_steps_per_second": 2.491, | |
| "step": 17252 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 8.81057268722467e-09, | |
| "loss": 0.4758, | |
| "step": 18160 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.44726473093032837, | |
| "eval_runtime": 40.5882, | |
| "eval_samples_per_second": 79.531, | |
| "eval_steps_per_second": 2.488, | |
| "step": 18160 | |
| } | |
| ], | |
| "max_steps": 18160, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.822589659829248e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |