| { | |
| "best_metric": 0.02586853690445423, | |
| "best_model_checkpoint": "autotrain-kno3k-fiasf/checkpoint-678", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 678, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 10.68883228302002, | |
| "learning_rate": 3.0882352941176475e-05, | |
| "loss": 0.8263, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 58.380653381347656, | |
| "learning_rate": 4.852459016393443e-05, | |
| "loss": 0.2512, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.1297217607498169, | |
| "learning_rate": 4.491803278688525e-05, | |
| "loss": 0.1081, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.23067115247249603, | |
| "learning_rate": 4.122950819672131e-05, | |
| "loss": 0.1776, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.0700414851307869, | |
| "learning_rate": 3.754098360655738e-05, | |
| "loss": 0.0775, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9844530816213215, | |
| "eval_f1_macro": 0.9790876623736272, | |
| "eval_f1_micro": 0.9844530816213215, | |
| "eval_f1_weighted": 0.9845629146174618, | |
| "eval_loss": 0.05487915500998497, | |
| "eval_precision_macro": 0.9728622841711356, | |
| "eval_precision_micro": 0.9844530816213215, | |
| "eval_precision_weighted": 0.9849390553498074, | |
| "eval_recall_macro": 0.9856721822982112, | |
| "eval_recall_micro": 0.9844530816213215, | |
| "eval_recall_weighted": 0.9844530816213215, | |
| "eval_runtime": 10.4593, | |
| "eval_samples_per_second": 172.191, | |
| "eval_steps_per_second": 10.804, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 0.03147374466061592, | |
| "learning_rate": 3.39344262295082e-05, | |
| "loss": 0.1134, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.15661084651947021, | |
| "learning_rate": 3.0245901639344264e-05, | |
| "loss": 0.069, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 0.04975114390254021, | |
| "learning_rate": 2.6557377049180327e-05, | |
| "loss": 0.0821, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 0.02913733199238777, | |
| "learning_rate": 2.2868852459016393e-05, | |
| "loss": 0.0636, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 0.03793076425790787, | |
| "learning_rate": 1.918032786885246e-05, | |
| "loss": 0.135, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9922265408106608, | |
| "eval_f1_macro": 0.9885732001605064, | |
| "eval_f1_micro": 0.9922265408106608, | |
| "eval_f1_weighted": 0.9922495320181636, | |
| "eval_loss": 0.0289547611027956, | |
| "eval_precision_macro": 0.9857016924464328, | |
| "eval_precision_micro": 0.9922265408106608, | |
| "eval_precision_weighted": 0.9923116897185307, | |
| "eval_recall_macro": 0.991509527641691, | |
| "eval_recall_micro": 0.9922265408106608, | |
| "eval_recall_weighted": 0.9922265408106608, | |
| "eval_runtime": 10.5141, | |
| "eval_samples_per_second": 171.294, | |
| "eval_steps_per_second": 10.747, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 23.56169891357422, | |
| "learning_rate": 1.5491803278688525e-05, | |
| "loss": 0.0508, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 0.0823952853679657, | |
| "learning_rate": 1.1803278688524591e-05, | |
| "loss": 0.0443, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.025019163265824318, | |
| "learning_rate": 8.114754098360657e-06, | |
| "loss": 0.1144, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 0.830748975276947, | |
| "learning_rate": 4.426229508196722e-06, | |
| "loss": 0.0624, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 1.670173168182373, | |
| "learning_rate": 7.377049180327869e-07, | |
| "loss": 0.131, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9938922820655192, | |
| "eval_f1_macro": 0.990657441189161, | |
| "eval_f1_micro": 0.9938922820655192, | |
| "eval_f1_weighted": 0.9939035669908675, | |
| "eval_loss": 0.02586853690445423, | |
| "eval_precision_macro": 0.9889770605232528, | |
| "eval_precision_micro": 0.9938922820655192, | |
| "eval_precision_weighted": 0.9939283176272281, | |
| "eval_recall_macro": 0.9923605914714783, | |
| "eval_recall_micro": 0.9938922820655192, | |
| "eval_recall_weighted": 0.9938922820655192, | |
| "eval_runtime": 10.4887, | |
| "eval_samples_per_second": 171.709, | |
| "eval_steps_per_second": 10.774, | |
| "step": 678 | |
| } | |
| ], | |
| "logging_steps": 45, | |
| "max_steps": 678, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.3599465039988531e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |