| { | |
| "best_metric": 0.06617802381515503, | |
| "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": 2.1988484859466553, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0245, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 2.0091118812561035, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 0.9455, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 2.0441346168518066, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 0.84, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.614025354385376, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 0.6962, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 4.7578816413879395, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 0.6628, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.082709550857544, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 0.5827, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 2.5298991203308105, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 0.5206, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.2223843336105347, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 0.3961, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.384169816970825, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 0.3538, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.968139171600342, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 0.366, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 2.683225393295288, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 0.332, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.5319671630859375, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 0.2441, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.454670429229736, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2797, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.2151002585887909, | |
| "eval_runtime": 0.5786, | |
| "eval_samples_per_second": 229.864, | |
| "eval_steps_per_second": 29.381, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.6131775379180908, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 0.1994, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.05777645111084, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2208, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.446441650390625, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.2071, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 0.513364851474762, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.2195, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 0.5959073901176453, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.2082, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 3.2837018966674805, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.164, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.47730451822280884, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.2523, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 0.6329728364944458, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.1814, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.2986414432525635, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.2185, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 2.1698338985443115, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.1693, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 4.337858200073242, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.2719, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 0.37527066469192505, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.1375, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3469255864620209, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1295, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.12535960972309113, | |
| "eval_runtime": 0.5604, | |
| "eval_samples_per_second": 237.348, | |
| "eval_steps_per_second": 30.338, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 0.44272086024284363, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.0902, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.3413867950439453, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.1415, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 24.360963821411133, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.1282, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 9.52725601196289, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.1328, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 4.562541961669922, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.1839, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.30225884914398193, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.1488, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 5.811612606048584, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.1863, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.1982144117355347, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.0703, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 0.34011611342430115, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.1382, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 8.479839324951172, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.1125, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 0.5773456692695618, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.1517, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.1365382671356201, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.1359, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.2860206365585327, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1402, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.09570833295583725, | |
| "eval_runtime": 0.5573, | |
| "eval_samples_per_second": 238.649, | |
| "eval_steps_per_second": 30.504, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.4339356422424316, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1388, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 5.613440990447998, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.1576, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 9.03783130645752, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.1395, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 0.23390205204486847, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.0787, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 9.974942207336426, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.1209, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 12.48496150970459, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.0825, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 6.40043306350708, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.0948, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 7.782913684844971, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.0743, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.22070997953414917, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.0965, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 0.21166592836380005, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.073, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.9372833371162415, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.1387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 0.20762138068675995, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.1324, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.30446386337280273, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.0819, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.06617802381515503, | |
| "eval_runtime": 0.5606, | |
| "eval_samples_per_second": 237.255, | |
| "eval_steps_per_second": 30.326, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 7.368893623352051, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.0984, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.6616872549057007, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.1661, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 0.34062761068344116, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.1551, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.3984853029251099, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.0507, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 0.2744806408882141, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.1306, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 0.2681213617324829, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.0827, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 9.644803047180176, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0777, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 2.2872049808502197, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.0891, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 5.324370861053467, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0875, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 0.2501937448978424, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.0845, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 0.2204637974500656, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0654, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.23217803239822388, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.0777, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.5837439298629761, | |
| "learning_rate": 0.0, | |
| "loss": 0.1172, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.08222565799951553, | |
| "eval_runtime": 0.5566, | |
| "eval_samples_per_second": 238.967, | |
| "eval_steps_per_second": 30.545, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.006371770595533e+17, | |
| "train_loss": 0.21963719349641067, | |
| "train_runtime": 49.864, | |
| "train_samples_per_second": 103.682, | |
| "train_steps_per_second": 13.035 | |
| } | |
| ], | |
| "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 | |
| } | |