| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.922928709055874, | |
| "eval_steps": 1866, | |
| "global_step": 18660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.992292870905588, | |
| "grad_norm": 0.014770383015275002, | |
| "learning_rate": 9.004287245444802e-06, | |
| "loss": 2.4856, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 5.992292870905588, | |
| "eval_accuracy": 0.023725514052804476, | |
| "eval_loss": 3.687168598175049, | |
| "eval_runtime": 32.1095, | |
| "eval_samples_per_second": 255.968, | |
| "eval_steps_per_second": 6.416, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 11.984585741811175, | |
| "grad_norm": 0.009874270297586918, | |
| "learning_rate": 8.004287245444803e-06, | |
| "loss": 2.4767, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 11.984585741811175, | |
| "eval_accuracy": 0.0243946952183964, | |
| "eval_loss": 3.6874239444732666, | |
| "eval_runtime": 31.8736, | |
| "eval_samples_per_second": 257.863, | |
| "eval_steps_per_second": 6.463, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 17.976878612716764, | |
| "grad_norm": 0.00965830311179161, | |
| "learning_rate": 7.004287245444802e-06, | |
| "loss": 2.4784, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 17.976878612716764, | |
| "eval_accuracy": 0.024698868475483634, | |
| "eval_loss": 3.6875760555267334, | |
| "eval_runtime": 32.1903, | |
| "eval_samples_per_second": 255.325, | |
| "eval_steps_per_second": 6.399, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 23.96917148362235, | |
| "grad_norm": 0.0134053910151124, | |
| "learning_rate": 6.004287245444802e-06, | |
| "loss": 2.4728, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 23.96917148362235, | |
| "eval_accuracy": 0.02506387638398832, | |
| "eval_loss": 3.6877822875976562, | |
| "eval_runtime": 31.794, | |
| "eval_samples_per_second": 258.508, | |
| "eval_steps_per_second": 6.479, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 29.961464354527937, | |
| "grad_norm": 0.024022573605179787, | |
| "learning_rate": 5.004287245444802e-06, | |
| "loss": 2.473, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.961464354527937, | |
| "eval_accuracy": 0.02552621973476092, | |
| "eval_loss": 3.687516927719116, | |
| "eval_runtime": 32.1071, | |
| "eval_samples_per_second": 255.987, | |
| "eval_steps_per_second": 6.416, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 35.95375722543353, | |
| "grad_norm": 2.473095655441284, | |
| "learning_rate": 4.0042872454448026e-06, | |
| "loss": 2.4621, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 35.95375722543353, | |
| "eval_accuracy": 0.026605020886563652, | |
| "eval_loss": 3.7371363639831543, | |
| "eval_runtime": 31.7929, | |
| "eval_samples_per_second": 258.517, | |
| "eval_steps_per_second": 6.479, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 41.946050096339114, | |
| "grad_norm": 8.784303665161133, | |
| "learning_rate": 3.004287245444802e-06, | |
| "loss": 2.3868, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 41.946050096339114, | |
| "eval_accuracy": 0.026941059913440982, | |
| "eval_loss": 3.8208398818969727, | |
| "eval_runtime": 31.9481, | |
| "eval_samples_per_second": 257.261, | |
| "eval_steps_per_second": 6.448, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 47.9383429672447, | |
| "grad_norm": 3.3316075801849365, | |
| "learning_rate": 2.004823151125402e-06, | |
| "loss": 2.3293, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 47.9383429672447, | |
| "eval_accuracy": 0.026752037960822483, | |
| "eval_loss": 3.8804972171783447, | |
| "eval_runtime": 31.8972, | |
| "eval_samples_per_second": 257.671, | |
| "eval_steps_per_second": 6.458, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 53.93063583815029, | |
| "grad_norm": 5.581373691558838, | |
| "learning_rate": 1.005359056806002e-06, | |
| "loss": 2.2917, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 53.93063583815029, | |
| "eval_accuracy": 0.02671317137797245, | |
| "eval_loss": 3.9712843894958496, | |
| "eval_runtime": 32.2575, | |
| "eval_samples_per_second": 254.794, | |
| "eval_steps_per_second": 6.386, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 59.922928709055874, | |
| "grad_norm": 7.843618392944336, | |
| "learning_rate": 5.894962486602358e-09, | |
| "loss": 2.2787, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.922928709055874, | |
| "eval_accuracy": 0.026633410390558464, | |
| "eval_loss": 4.026557922363281, | |
| "eval_runtime": 31.9343, | |
| "eval_samples_per_second": 257.372, | |
| "eval_steps_per_second": 6.451, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.922928709055874, | |
| "step": 18660, | |
| "total_flos": 1.4768051860614912e+18, | |
| "train_loss": 2.41349681596664, | |
| "train_runtime": 33441.0012, | |
| "train_samples_per_second": 111.698, | |
| "train_steps_per_second": 0.558 | |
| } | |
| ], | |
| "logging_steps": 1866, | |
| "max_steps": 18660, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4768051860614912e+18, | |
| "train_batch_size": 40, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |