| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.117437722419929, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.650997150997152e-05, | |
| "loss": 1.5302, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.295584045584046e-05, | |
| "loss": 1.3916, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 8.939458689458689e-05, | |
| "loss": 1.2024, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 8.583333333333334e-05, | |
| "loss": 1.0846, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 8.227207977207978e-05, | |
| "loss": 1.0088, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 7.871082621082622e-05, | |
| "loss": 0.9088, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.514957264957265e-05, | |
| "loss": 0.8388, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 7.158831908831909e-05, | |
| "loss": 0.7937, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 6.802706552706552e-05, | |
| "loss": 0.7683, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 6.446581196581197e-05, | |
| "loss": 0.734, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 6.090455840455841e-05, | |
| "loss": 0.7031, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.7343304843304845e-05, | |
| "loss": 0.6465, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 5.378205128205128e-05, | |
| "loss": 0.6236, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 5.022079772079772e-05, | |
| "loss": 0.6151, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4.665954415954416e-05, | |
| "loss": 0.5924, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 4.30982905982906e-05, | |
| "loss": 0.5918, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 3.9537037037037036e-05, | |
| "loss": 0.5756, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 3.597578347578348e-05, | |
| "loss": 0.5302, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 3.241452991452992e-05, | |
| "loss": 0.5245, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.8853276353276353e-05, | |
| "loss": 0.5219, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_bleu": 9.2289, | |
| "eval_gen_len": 36.3363, | |
| "eval_loss": 0.7101184129714966, | |
| "eval_runtime": 335.0095, | |
| "eval_samples_per_second": 11.892, | |
| "eval_steps_per_second": 0.373, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 2.5292022792022797e-05, | |
| "loss": 0.4957, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 2.173076923076923e-05, | |
| "loss": 0.4974, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 1.817663817663818e-05, | |
| "loss": 0.4834, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 1.4615384615384617e-05, | |
| "loss": 0.4756, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 1.1054131054131055e-05, | |
| "loss": 0.4713, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 7.492877492877494e-06, | |
| "loss": 0.4682, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 3.9316239316239315e-06, | |
| "loss": 0.4713, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 3.7749287749287755e-07, | |
| "loss": 0.4758, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0, | |
| "loss": 0.4562, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 0.0, | |
| "loss": 0.457, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 0.0, | |
| "loss": 0.4522, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.0, | |
| "loss": 0.4607, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 0.0, | |
| "loss": 0.4508, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.0, | |
| "loss": 0.4547, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 0.0, | |
| "loss": 0.4495, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 0.0, | |
| "loss": 0.4565, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 0.0, | |
| "loss": 0.4553, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 0.0, | |
| "loss": 0.4521, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.0, | |
| "loss": 0.4621, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 0.0, | |
| "loss": 0.4585, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "eval_bleu": 9.9179, | |
| "eval_gen_len": 36.9039, | |
| "eval_loss": 0.6802528500556946, | |
| "eval_runtime": 335.8822, | |
| "eval_samples_per_second": 11.861, | |
| "eval_steps_per_second": 0.372, | |
| "step": 20000 | |
| } | |
| ], | |
| "max_steps": 28100, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.338326302831411e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |