| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20833333333333334, | |
| "grad_norm": 0.03092513605952263, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.4044, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 0.03055603615939617, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.3836, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.023527173325419426, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0017, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.02501656673848629, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0008, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.968421052631579, | |
| "eval_f1": 0.9770992366412214, | |
| "eval_loss": 0.23919111490249634, | |
| "eval_precision": 0.9696969696969697, | |
| "eval_recall": 0.9846153846153847, | |
| "eval_runtime": 2.2913, | |
| "eval_samples_per_second": 41.461, | |
| "eval_steps_per_second": 2.619, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0416666666666667, | |
| "grad_norm": 0.027175450697541237, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0007, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.017659997567534447, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0006, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4583333333333333, | |
| "grad_norm": 0.01459481567144394, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.0007, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.011688518337905407, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0028, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.08688002079725266, | |
| "learning_rate": 9e-06, | |
| "loss": 0.0006, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9052631578947369, | |
| "eval_f1": 0.9312977099236641, | |
| "eval_loss": 0.7143413424491882, | |
| "eval_precision": 0.9242424242424242, | |
| "eval_recall": 0.9384615384615385, | |
| "eval_runtime": 2.2819, | |
| "eval_samples_per_second": 41.632, | |
| "eval_steps_per_second": 2.629, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 0.029978642240166664, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1005, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.2916666666666665, | |
| "grad_norm": 0.010506043210625648, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.0003, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 413.86444091796875, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1257, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.7083333333333335, | |
| "grad_norm": 0.022496210411190987, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.0018, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.9166666666666665, | |
| "grad_norm": 4.833926677703857, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.0823, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_f1": 0.962406015037594, | |
| "eval_loss": 0.2848361134529114, | |
| "eval_precision": 0.9411764705882353, | |
| "eval_recall": 0.9846153846153847, | |
| "eval_runtime": 2.3284, | |
| "eval_samples_per_second": 40.8, | |
| "eval_steps_per_second": 2.577, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 133.06631469726562, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0298, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 5.606410503387451, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1977, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.5416666666666665, | |
| "grad_norm": 1.3438758850097656, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.0345, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 219.62388610839844, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.2788, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.9583333333333335, | |
| "grad_norm": 0.02319713868200779, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.0778, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.9285714285714286, | |
| "eval_loss": 0.73874431848526, | |
| "eval_precision": 0.8666666666666667, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 2.3292, | |
| "eval_samples_per_second": 40.786, | |
| "eval_steps_per_second": 2.576, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 0.07177528738975525, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2188, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 0.025927864015102386, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.0018, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.583333333333333, | |
| "grad_norm": 0.25771522521972656, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.2547, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.791666666666667, | |
| "grad_norm": 0.1835576444864273, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.1684, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.051910556852817535, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.1683, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9263157894736842, | |
| "eval_f1": 0.9465648854961831, | |
| "eval_loss": 0.482939213514328, | |
| "eval_precision": 0.9393939393939394, | |
| "eval_recall": 0.9538461538461539, | |
| "eval_runtime": 2.321, | |
| "eval_samples_per_second": 40.931, | |
| "eval_steps_per_second": 2.585, | |
| "step": 240 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 442667397580800.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |