| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.358400662937642, | |
| "eval_steps": 50000, | |
| "global_step": 650000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 1.1412, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.011365294456482, | |
| "eval_runtime": 2006.5534, | |
| "eval_samples_per_second": 111.193, | |
| "eval_steps_per_second": 1.738, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.992035346050262e-05, | |
| "loss": 1.0123, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_loss": 0.9693423509597778, | |
| "eval_runtime": 1995.4615, | |
| "eval_samples_per_second": 111.811, | |
| "eval_steps_per_second": 1.747, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 4.992034708610221e-05, | |
| "loss": 0.9754, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_loss": 0.9472731351852417, | |
| "eval_runtime": 1997.2157, | |
| "eval_samples_per_second": 111.713, | |
| "eval_steps_per_second": 1.746, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.992034549250211e-05, | |
| "loss": 0.9539, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "eval_loss": 0.9325647354125977, | |
| "eval_runtime": 2015.1208, | |
| "eval_samples_per_second": 110.72, | |
| "eval_steps_per_second": 1.73, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 4.992035027330242e-05, | |
| "loss": 0.9387, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "eval_loss": 0.9212433099746704, | |
| "eval_runtime": 2011.4269, | |
| "eval_samples_per_second": 110.924, | |
| "eval_steps_per_second": 1.734, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 4.992035027330242e-05, | |
| "loss": 0.9243, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_loss": 0.9138051271438599, | |
| "eval_runtime": 2011.3973, | |
| "eval_samples_per_second": 110.925, | |
| "eval_steps_per_second": 1.734, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.9144, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "eval_loss": 0.9093130826950073, | |
| "eval_runtime": 1998.4573, | |
| "eval_samples_per_second": 111.644, | |
| "eval_steps_per_second": 1.745, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 0.906, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "eval_loss": 0.9041373133659363, | |
| "eval_runtime": 1998.7351, | |
| "eval_samples_per_second": 111.628, | |
| "eval_steps_per_second": 1.745, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 0.8994, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "eval_loss": 0.9003444910049438, | |
| "eval_runtime": 1982.6092, | |
| "eval_samples_per_second": 112.536, | |
| "eval_steps_per_second": 1.759, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 4.992035186690252e-05, | |
| "loss": 0.8933, | |
| "step": 500000 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "eval_loss": 0.8956149220466614, | |
| "eval_runtime": 2002.7479, | |
| "eval_samples_per_second": 111.404, | |
| "eval_steps_per_second": 1.741, | |
| "step": 500000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 4.9920355054102726e-05, | |
| "loss": 0.8856, | |
| "step": 550000 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "eval_loss": 0.8930546045303345, | |
| "eval_runtime": 1996.5839, | |
| "eval_samples_per_second": 111.748, | |
| "eval_steps_per_second": 1.746, | |
| "step": 550000 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 4.992035346050262e-05, | |
| "loss": 0.8802, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "eval_loss": 0.89084392786026, | |
| "eval_runtime": 1991.4984, | |
| "eval_samples_per_second": 112.034, | |
| "eval_steps_per_second": 1.751, | |
| "step": 600000 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "learning_rate": 4.992035346050262e-05, | |
| "loss": 0.8763, | |
| "step": 650000 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "eval_loss": 0.8895950317382812, | |
| "eval_runtime": 2006.0064, | |
| "eval_samples_per_second": 111.223, | |
| "eval_steps_per_second": 1.738, | |
| "step": 650000 | |
| } | |
| ], | |
| "logging_steps": 50000, | |
| "max_steps": 31375500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 500, | |
| "save_steps": 50000, | |
| "total_flos": 1.3587153002496e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |