| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.87154784842646, | |
| "eval_steps": 3110, | |
| "global_step": 31100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.987154784842646, | |
| "grad_norm": 0.005527158733457327, | |
| "learning_rate": 9e-06, | |
| "loss": 0.0008, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 9.987154784842646, | |
| "eval_accuracy": 0.050006083465141744, | |
| "eval_loss": 3.0008227825164795, | |
| "eval_runtime": 32.0118, | |
| "eval_samples_per_second": 256.749, | |
| "eval_steps_per_second": 12.839, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 19.974309569685293, | |
| "grad_norm": 0.0, | |
| "learning_rate": 8.00032154340836e-06, | |
| "loss": 0.0007, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 19.974309569685293, | |
| "eval_accuracy": 0.050066918116559195, | |
| "eval_loss": 2.997060775756836, | |
| "eval_runtime": 32.3792, | |
| "eval_samples_per_second": 253.836, | |
| "eval_steps_per_second": 12.693, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 29.961464354527937, | |
| "grad_norm": 0.0, | |
| "learning_rate": 7.000643086816721e-06, | |
| "loss": 0.0007, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.961464354527937, | |
| "eval_accuracy": 0.05008719633369834, | |
| "eval_loss": 2.997131586074829, | |
| "eval_runtime": 32.4433, | |
| "eval_samples_per_second": 253.335, | |
| "eval_steps_per_second": 12.668, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 39.948619139370585, | |
| "grad_norm": 0.0021410025656223297, | |
| "learning_rate": 6.00128617363344e-06, | |
| "loss": 0.0007, | |
| "step": 12440 | |
| }, | |
| { | |
| "epoch": 39.948619139370585, | |
| "eval_accuracy": 0.050006083465141744, | |
| "eval_loss": 2.998797655105591, | |
| "eval_runtime": 32.4005, | |
| "eval_samples_per_second": 253.669, | |
| "eval_steps_per_second": 12.685, | |
| "step": 12440 | |
| }, | |
| { | |
| "epoch": 49.93577392421323, | |
| "grad_norm": 0.0, | |
| "learning_rate": 5.001607717041801e-06, | |
| "loss": 0.0007, | |
| "step": 15550 | |
| }, | |
| { | |
| "epoch": 49.93577392421323, | |
| "eval_accuracy": 0.04990874802287383, | |
| "eval_loss": 2.9968416690826416, | |
| "eval_runtime": 32.3879, | |
| "eval_samples_per_second": 253.768, | |
| "eval_steps_per_second": 12.69, | |
| "step": 15550 | |
| }, | |
| { | |
| "epoch": 59.922928709055874, | |
| "grad_norm": 0.0, | |
| "learning_rate": 4.002250803858521e-06, | |
| "loss": 0.0007, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.922928709055874, | |
| "eval_accuracy": 0.050168309202254936, | |
| "eval_loss": 2.9966351985931396, | |
| "eval_runtime": 32.4157, | |
| "eval_samples_per_second": 253.55, | |
| "eval_steps_per_second": 12.679, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 69.91008349389853, | |
| "grad_norm": 0.0, | |
| "learning_rate": 3.0022508038585212e-06, | |
| "loss": 0.0007, | |
| "step": 21770 | |
| }, | |
| { | |
| "epoch": 69.91008349389853, | |
| "eval_accuracy": 0.050301566057740776, | |
| "eval_loss": 2.9960734844207764, | |
| "eval_runtime": 32.1842, | |
| "eval_samples_per_second": 255.374, | |
| "eval_steps_per_second": 12.77, | |
| "step": 21770 | |
| }, | |
| { | |
| "epoch": 79.89723827874117, | |
| "grad_norm": 0.0, | |
| "learning_rate": 2.002893890675241e-06, | |
| "loss": 0.0007, | |
| "step": 24880 | |
| }, | |
| { | |
| "epoch": 79.89723827874117, | |
| "eval_accuracy": 0.05029504805937462, | |
| "eval_loss": 2.9966866970062256, | |
| "eval_runtime": 32.39, | |
| "eval_samples_per_second": 253.751, | |
| "eval_steps_per_second": 12.689, | |
| "step": 24880 | |
| }, | |
| { | |
| "epoch": 89.88439306358381, | |
| "grad_norm": 0.0, | |
| "learning_rate": 1.0032154340836014e-06, | |
| "loss": 0.0007, | |
| "step": 27990 | |
| }, | |
| { | |
| "epoch": 89.88439306358381, | |
| "eval_accuracy": 0.050303497316515934, | |
| "eval_loss": 2.996614694595337, | |
| "eval_runtime": 32.3491, | |
| "eval_samples_per_second": 254.072, | |
| "eval_steps_per_second": 12.705, | |
| "step": 27990 | |
| }, | |
| { | |
| "epoch": 99.87154784842646, | |
| "grad_norm": 0.0, | |
| "learning_rate": 3.858520900321544e-09, | |
| "loss": 0.0007, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 99.87154784842646, | |
| "eval_accuracy": 0.05015208662854362, | |
| "eval_loss": 2.9962856769561768, | |
| "eval_runtime": 32.3824, | |
| "eval_samples_per_second": 253.811, | |
| "eval_steps_per_second": 12.692, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 99.87154784842646, | |
| "step": 31100, | |
| "total_flos": 2.458812304495296e+18, | |
| "train_loss": 0.000715851561432866, | |
| "train_runtime": 54755.5148, | |
| "train_samples_per_second": 113.696, | |
| "train_steps_per_second": 0.568 | |
| } | |
| ], | |
| "logging_steps": 3110, | |
| "max_steps": 31100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 2.458812304495296e+18, | |
| "train_batch_size": 40, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |