| { | |
| "best_metric": 0.06386490911245346, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-520", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0246, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 0.9458, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 0.8442, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 0.7014, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 0.6649, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 0.5867, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 0.5239, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 0.3946, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 0.3543, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 0.3679, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 0.3318, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 0.2449, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2804, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.21983224153518677, | |
| "eval_runtime": 5.6144, | |
| "eval_samples_per_second": 23.689, | |
| "eval_steps_per_second": 3.028, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 0.2006, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2294, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.1983, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.2172, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.2044, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.1594, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.2487, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.1816, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.2222, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.149, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.2807, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.1358, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1304, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.12584400177001953, | |
| "eval_runtime": 5.8683, | |
| "eval_samples_per_second": 22.664, | |
| "eval_steps_per_second": 2.897, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.093, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.1379, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.131, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.1421, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.1665, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.1529, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.1775, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.075, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.1445, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.1108, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.1449, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.1261, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1388, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.09265834838151932, | |
| "eval_runtime": 6.036, | |
| "eval_samples_per_second": 22.034, | |
| "eval_steps_per_second": 2.816, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1353, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.157, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.1425, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.0786, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.1265, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.0785, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.0962, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.0718, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.087, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.0697, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1455, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.1316, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.088, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.06386490911245346, | |
| "eval_runtime": 6.0283, | |
| "eval_samples_per_second": 22.063, | |
| "eval_steps_per_second": 2.82, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.099, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.1589, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.1542, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0504, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.1352, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0809, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0653, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.1116, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0909, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0731, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0643, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.0783, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1175, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.08365032821893692, | |
| "eval_runtime": 6.1202, | |
| "eval_samples_per_second": 21.731, | |
| "eval_steps_per_second": 2.778, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_loss": 0.21925977853628306, | |
| "train_runtime": 631.9353, | |
| "train_samples_per_second": 8.181, | |
| "train_steps_per_second": 1.029 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |