| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 3000, | |
| "global_step": 44343, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.76544212164265e-05, | |
| "grad_norm": 13.6875, | |
| "learning_rate": 0.0002999932345578783, | |
| "loss": 4.8125, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.20296326364927947, | |
| "grad_norm": 0.9453125, | |
| "learning_rate": 0.000279703673635072, | |
| "loss": 3.1476, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.20296326364927947, | |
| "eval_loss": 2.6158535480499268, | |
| "eval_runtime": 54.7052, | |
| "eval_samples_per_second": 1725.705, | |
| "eval_steps_per_second": 6.745, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.40592652729855894, | |
| "grad_norm": 0.98046875, | |
| "learning_rate": 0.0002594073472701441, | |
| "loss": 3.0318, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.40592652729855894, | |
| "eval_loss": 2.588202953338623, | |
| "eval_runtime": 54.5614, | |
| "eval_samples_per_second": 1730.252, | |
| "eval_steps_per_second": 6.763, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6088897909478385, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 0.0002391110209052161, | |
| "loss": 3.0079, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6088897909478385, | |
| "eval_loss": 2.5750339031219482, | |
| "eval_runtime": 54.4102, | |
| "eval_samples_per_second": 1735.062, | |
| "eval_steps_per_second": 6.782, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8118530545971179, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 0.0002188146945402882, | |
| "loss": 3.0, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.8118530545971179, | |
| "eval_loss": 2.5675389766693115, | |
| "eval_runtime": 54.4389, | |
| "eval_samples_per_second": 1734.145, | |
| "eval_steps_per_second": 6.778, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0148163182463974, | |
| "grad_norm": 0.99609375, | |
| "learning_rate": 0.00019851836817536025, | |
| "loss": 2.9932, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.0148163182463974, | |
| "eval_loss": 2.564955949783325, | |
| "eval_runtime": 54.5924, | |
| "eval_samples_per_second": 1729.27, | |
| "eval_steps_per_second": 6.759, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.217779581895677, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 0.0001782220418104323, | |
| "loss": 2.9891, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.217779581895677, | |
| "eval_loss": 2.5631775856018066, | |
| "eval_runtime": 54.5754, | |
| "eval_samples_per_second": 1729.81, | |
| "eval_steps_per_second": 6.761, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.4207428455449564, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.00015792571544550436, | |
| "loss": 2.9869, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.4207428455449564, | |
| "eval_loss": 2.560298204421997, | |
| "eval_runtime": 54.5785, | |
| "eval_samples_per_second": 1729.709, | |
| "eval_steps_per_second": 6.761, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.6237061091942357, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 0.0001376293890805764, | |
| "loss": 2.9846, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.6237061091942357, | |
| "eval_loss": 2.5608062744140625, | |
| "eval_runtime": 54.7118, | |
| "eval_samples_per_second": 1725.495, | |
| "eval_steps_per_second": 6.744, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.8266693728435153, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.00011733306271564845, | |
| "loss": 2.9831, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.8266693728435153, | |
| "eval_loss": 2.559070110321045, | |
| "eval_runtime": 54.5867, | |
| "eval_samples_per_second": 1729.449, | |
| "eval_steps_per_second": 6.76, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.029632636492795, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 9.703673635072052e-05, | |
| "loss": 2.9832, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.029632636492795, | |
| "eval_loss": 2.559154748916626, | |
| "eval_runtime": 54.5764, | |
| "eval_samples_per_second": 1729.778, | |
| "eval_steps_per_second": 6.761, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.232595900142074, | |
| "grad_norm": 1.1953125, | |
| "learning_rate": 7.674040998579256e-05, | |
| "loss": 2.9811, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.232595900142074, | |
| "eval_loss": 2.559070110321045, | |
| "eval_runtime": 54.4327, | |
| "eval_samples_per_second": 1734.344, | |
| "eval_steps_per_second": 6.779, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.435559163791354, | |
| "grad_norm": 1.203125, | |
| "learning_rate": 5.644408362086462e-05, | |
| "loss": 2.9841, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.435559163791354, | |
| "eval_loss": 2.559366464614868, | |
| "eval_runtime": 54.4466, | |
| "eval_samples_per_second": 1733.901, | |
| "eval_steps_per_second": 6.777, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.638522427440633, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 3.614775725593667e-05, | |
| "loss": 2.9836, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.638522427440633, | |
| "eval_loss": 2.558900833129883, | |
| "eval_runtime": 54.4453, | |
| "eval_samples_per_second": 1733.941, | |
| "eval_steps_per_second": 6.777, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.841485691089913, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 1.5851430891008727e-05, | |
| "loss": 2.9827, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.841485691089913, | |
| "eval_loss": 2.559112548828125, | |
| "eval_runtime": 54.4302, | |
| "eval_samples_per_second": 1734.422, | |
| "eval_steps_per_second": 6.779, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 44343, | |
| "total_flos": 3.8956704097296384e+17, | |
| "train_loss": 3.0017653575536163, | |
| "train_runtime": 20075.3447, | |
| "train_samples_per_second": 565.443, | |
| "train_steps_per_second": 2.209 | |
| } | |
| ], | |
| "logging_steps": 3000, | |
| "max_steps": 44343, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.8956704097296384e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |