| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 36816, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 5.431830526887561e-06, | |
| "loss": 0.9928, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.0863661053775122e-05, | |
| "loss": 0.7822, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.6295491580662687e-05, | |
| "loss": 0.7134, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.9808052151868172e-05, | |
| "loss": 0.6789, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.9204442566547958e-05, | |
| "loss": 0.647, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.8600832981227744e-05, | |
| "loss": 0.6252, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.7997223395907527e-05, | |
| "loss": 0.6084, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.7393613810587313e-05, | |
| "loss": 0.5935, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.67900042252671e-05, | |
| "loss": 0.5812, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.618639463994688e-05, | |
| "loss": 0.5643, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.5582785054626667e-05, | |
| "loss": 0.5629, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.4979175469306453e-05, | |
| "loss": 0.556, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5368458032608032, | |
| "eval_runtime": 2.6095, | |
| "eval_samples_per_second": 954.204, | |
| "eval_steps_per_second": 29.891, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.437556588398624e-05, | |
| "loss": 0.4938, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.3771956298666025e-05, | |
| "loss": 0.4773, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.316834671334581e-05, | |
| "loss": 0.4797, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.2564737128025594e-05, | |
| "loss": 0.4754, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.196112754270538e-05, | |
| "loss": 0.4692, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.1357517957385164e-05, | |
| "loss": 0.4682, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.0753908372064949e-05, | |
| "loss": 0.4632, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.0150298786744735e-05, | |
| "loss": 0.4611, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 9.546689201424519e-06, | |
| "loss": 0.4507, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 8.943079616104305e-06, | |
| "loss": 0.4487, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 8.33947003078409e-06, | |
| "loss": 0.4493, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 7.735860445463874e-06, | |
| "loss": 0.4513, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.4784027636051178, | |
| "eval_runtime": 2.5508, | |
| "eval_samples_per_second": 976.183, | |
| "eval_steps_per_second": 30.579, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 7.1322508601436605e-06, | |
| "loss": 0.4059, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 6.528641274823445e-06, | |
| "loss": 0.3585, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.92503168950323e-06, | |
| "loss": 0.3568, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 5.321422104183015e-06, | |
| "loss": 0.3566, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 4.7178125188628e-06, | |
| "loss": 0.3484, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.114202933542585e-06, | |
| "loss": 0.3527, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3.51059334822237e-06, | |
| "loss": 0.3561, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.9069837629021554e-06, | |
| "loss": 0.3512, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.30337417758194e-06, | |
| "loss": 0.3549, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.6997645922617255e-06, | |
| "loss": 0.3555, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.0961550069415102e-06, | |
| "loss": 0.3396, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.925454216212955e-07, | |
| "loss": 0.3422, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.49958956241607666, | |
| "eval_runtime": 2.5304, | |
| "eval_samples_per_second": 984.052, | |
| "eval_steps_per_second": 30.826, | |
| "step": 36816 | |
| } | |
| ], | |
| "max_steps": 36816, | |
| "num_train_epochs": 3, | |
| "total_flos": 7.749426953899213e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |