| { | |
| "best_metric": 0.06496671587228775, | |
| "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": 0.07692307692307693, | |
| "grad_norm": 2.1983582973480225, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0245, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.8467624187469482, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 0.9453, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 2.103004217147827, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 0.8407, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.6212635040283203, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 0.6962, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 4.0500969886779785, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 0.6627, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.1369540691375732, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 0.5824, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 1.9217969179153442, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 0.5207, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.2160992622375488, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 0.3929, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.5192856788635254, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 0.3531, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.9285073280334473, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 0.3694, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 2.098353147506714, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 0.3324, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.5526487827301025, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 0.2447, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.012009143829346, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2838, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.21691447496414185, | |
| "eval_runtime": 1.8915, | |
| "eval_samples_per_second": 70.313, | |
| "eval_steps_per_second": 8.987, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.6030009984970093, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 0.1997, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.190946578979492, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2249, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.4254164695739746, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.2047, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 0.5190562605857849, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.2217, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 0.6067346334457397, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.2069, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 3.877462387084961, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.1627, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.4851893484592438, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.2515, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 0.7414268851280212, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.1788, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.4538869857788086, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.2098, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 1.7667008638381958, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.1678, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 4.910299777984619, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.2644, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 0.383217453956604, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.1394, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3504723310470581, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1324, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.13457302749156952, | |
| "eval_runtime": 2.0055, | |
| "eval_samples_per_second": 66.318, | |
| "eval_steps_per_second": 8.477, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 0.5011544227600098, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.0904, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.3368860185146332, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.1434, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 6.839877128601074, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.124, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 8.27059555053711, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.1441, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 4.772457599639893, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.1723, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.3270654082298279, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.1571, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 6.340713024139404, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.1882, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.2030677795410156, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.0736, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 0.3556363880634308, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.1333, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 10.645792961120605, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.1295, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 0.30723875761032104, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.1506, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.1151694059371948, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.1301, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.29431405663490295, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.143, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.09983458369970322, | |
| "eval_runtime": 2.4088, | |
| "eval_samples_per_second": 55.214, | |
| "eval_steps_per_second": 7.057, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.359541893005371, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1319, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 6.216188907623291, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.1523, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 9.072651863098145, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.1394, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 0.23217681050300598, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.0776, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 8.235204696655273, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.1129, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 12.133755683898926, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.0812, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 5.274909496307373, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.0968, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 8.703495025634766, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.0805, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.21897560358047485, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.0964, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 0.21305297315120697, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.075, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.9238463044166565, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1348, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 0.21227774024009705, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.127, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.30383384227752686, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0833, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.06496671587228775, | |
| "eval_runtime": 2.4064, | |
| "eval_samples_per_second": 55.27, | |
| "eval_steps_per_second": 7.065, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 5.725286960601807, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.0969, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.5072094202041626, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.1634, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 0.3428334891796112, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.1552, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.1749656200408936, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0503, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 0.32822975516319275, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.1335, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 0.2676811218261719, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0836, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 8.637232780456543, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0739, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 2.967637777328491, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0973, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 4.9968953132629395, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0857, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 0.2506353259086609, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0831, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 0.21034254133701324, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.066, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.23136074841022491, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.0781, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6315775513648987, | |
| "learning_rate": 0.0, | |
| "loss": 0.1099, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.0823998898267746, | |
| "eval_runtime": 2.4012, | |
| "eval_samples_per_second": 55.389, | |
| "eval_steps_per_second": 7.08, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_loss": 0.0, | |
| "train_runtime": 0.178, | |
| "train_samples_per_second": 29040.514, | |
| "train_steps_per_second": 3651.128 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 4.006371770595533e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |