| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3074691805656273, | |
| "global_step": 450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.75e-05, | |
| "loss": 12.1562, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 7.125e-05, | |
| "loss": 8.7679, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 7.398952095808383e-05, | |
| "loss": 5.3683, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 7.286676646706586e-05, | |
| "loss": 4.3219, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 7.17440119760479e-05, | |
| "loss": 3.7182, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 3.836604595184326, | |
| "eval_runtime": 133.4846, | |
| "eval_samples_per_second": 34.611, | |
| "eval_steps_per_second": 4.33, | |
| "eval_wer": 1.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 7.062125748502993e-05, | |
| "loss": 3.478, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 6.949850299401197e-05, | |
| "loss": 3.4492, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 6.837574850299401e-05, | |
| "loss": 3.3928, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.725299401197604e-05, | |
| "loss": 3.3183, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 6.613023952095809e-05, | |
| "loss": 3.2075, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 3.258362293243408, | |
| "eval_runtime": 126.6078, | |
| "eval_samples_per_second": 36.491, | |
| "eval_steps_per_second": 4.565, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.500748502994012e-05, | |
| "loss": 3.14, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.388473053892215e-05, | |
| "loss": 3.1281, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.276197604790418e-05, | |
| "loss": 3.0987, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.163922155688622e-05, | |
| "loss": 3.1003, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 6.0516467065868256e-05, | |
| "loss": 3.0922, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 3.127869129180908, | |
| "eval_runtime": 126.3837, | |
| "eval_samples_per_second": 36.555, | |
| "eval_steps_per_second": 4.573, | |
| "eval_wer": 1.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.9393712574850293e-05, | |
| "loss": 3.0588, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 5.827095808383233e-05, | |
| "loss": 3.0477, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 5.714820359281436e-05, | |
| "loss": 3.045, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 5.602544910179641e-05, | |
| "loss": 3.0439, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 5.490269461077844e-05, | |
| "loss": 3.0846, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 3.079519271850586, | |
| "eval_runtime": 125.7215, | |
| "eval_samples_per_second": 36.748, | |
| "eval_steps_per_second": 4.597, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 5.3779940119760477e-05, | |
| "loss": 3.0512, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 5.265718562874251e-05, | |
| "loss": 3.0143, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 5.1534431137724546e-05, | |
| "loss": 3.0387, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 5.0411676646706584e-05, | |
| "loss": 3.0311, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.9288922155688615e-05, | |
| "loss": 3.0417, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 3.069390058517456, | |
| "eval_runtime": 125.7339, | |
| "eval_samples_per_second": 36.744, | |
| "eval_steps_per_second": 4.597, | |
| "eval_wer": 1.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.816616766467066e-05, | |
| "loss": 3.0219, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.704341317365269e-05, | |
| "loss": 3.0194, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.592065868263473e-05, | |
| "loss": 2.9974, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.479790419161676e-05, | |
| "loss": 2.9996, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.36751497005988e-05, | |
| "loss": 3.0016, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 3.0347490310668945, | |
| "eval_runtime": 132.3039, | |
| "eval_samples_per_second": 34.92, | |
| "eval_steps_per_second": 4.369, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.255239520958083e-05, | |
| "loss": 3.0052, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.142964071856287e-05, | |
| "loss": 2.9826, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.030688622754491e-05, | |
| "loss": 2.9747, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.918413173652694e-05, | |
| "loss": 2.9617, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3.806137724550898e-05, | |
| "loss": 3.2053, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.984886407852173, | |
| "eval_runtime": 149.1508, | |
| "eval_samples_per_second": 30.975, | |
| "eval_steps_per_second": 3.875, | |
| "eval_wer": 1.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.693862275449102e-05, | |
| "loss": 2.9665, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3.581586826347305e-05, | |
| "loss": 2.9641, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 3.469311377245509e-05, | |
| "loss": 2.9484, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.3570359281437126e-05, | |
| "loss": 2.9494, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 3.244760479041916e-05, | |
| "loss": 2.9698, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 2.989494562149048, | |
| "eval_runtime": 150.6903, | |
| "eval_samples_per_second": 30.659, | |
| "eval_steps_per_second": 3.836, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.1324850299401195e-05, | |
| "loss": 2.9664, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.020209580838323e-05, | |
| "loss": 2.9494, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.9079341317365265e-05, | |
| "loss": 2.935, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.7956586826347306e-05, | |
| "loss": 2.9397, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.683383233532934e-05, | |
| "loss": 2.9485, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.9584460258483887, | |
| "eval_runtime": 140.6358, | |
| "eval_samples_per_second": 32.851, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.0, | |
| "step": 450 | |
| } | |
| ], | |
| "max_steps": 688, | |
| "num_train_epochs": 2, | |
| "total_flos": 1.336010418574825e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |