| { | |
| "best_metric": 0.8212058967880319, | |
| "best_model_checkpoint": "./outputs/facebook/wav2vec2-xls-r-1b/checkpoint-6716", | |
| "epoch": 29.969293756397136, | |
| "eval_steps": 500, | |
| "global_step": 18300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.07e-05, | |
| "loss": 4.5214, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.5036747455596924, | |
| "eval_runtime": 53.3547, | |
| "eval_samples_per_second": 139.182, | |
| "eval_steps_per_second": 4.367, | |
| "eval_wer": 0.9721338592696271, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.873988439306359e-05, | |
| "loss": 0.5588, | |
| "step": 1221 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.375758171081543, | |
| "eval_runtime": 52.8335, | |
| "eval_samples_per_second": 140.555, | |
| "eval_steps_per_second": 4.41, | |
| "eval_wer": 0.9198734844491302, | |
| "step": 1221 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.521387283236995e-05, | |
| "loss": 0.4547, | |
| "step": 1831 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.2466094493865967, | |
| "eval_runtime": 53.1764, | |
| "eval_samples_per_second": 139.648, | |
| "eval_steps_per_second": 4.382, | |
| "eval_wer": 0.9418864631996073, | |
| "step": 1831 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.16820809248555e-05, | |
| "loss": 0.4025, | |
| "step": 2442 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.2312777042388916, | |
| "eval_runtime": 54.1165, | |
| "eval_samples_per_second": 137.223, | |
| "eval_steps_per_second": 4.306, | |
| "eval_wer": 0.8876083834729973, | |
| "step": 2442 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.815606936416185e-05, | |
| "loss": 0.3789, | |
| "step": 3053 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.244392156600952, | |
| "eval_runtime": 52.4474, | |
| "eval_samples_per_second": 141.589, | |
| "eval_steps_per_second": 4.443, | |
| "eval_wer": 0.8844636722229291, | |
| "step": 3053 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.463005780346821e-05, | |
| "loss": 0.3542, | |
| "step": 3663 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.9893066883087158, | |
| "eval_runtime": 54.83, | |
| "eval_samples_per_second": 135.437, | |
| "eval_steps_per_second": 4.249, | |
| "eval_wer": 0.8913711304600731, | |
| "step": 3663 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 8.109826589595376e-05, | |
| "loss": 0.3383, | |
| "step": 4274 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.3484952449798584, | |
| "eval_runtime": 53.4092, | |
| "eval_samples_per_second": 139.04, | |
| "eval_steps_per_second": 4.363, | |
| "eval_wer": 0.8356752040426808, | |
| "step": 4274 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 7.756647398843931e-05, | |
| "loss": 0.3239, | |
| "step": 4885 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.98879075050354, | |
| "eval_runtime": 53.1444, | |
| "eval_samples_per_second": 139.732, | |
| "eval_steps_per_second": 4.384, | |
| "eval_wer": 0.8775743915074619, | |
| "step": 4885 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 7.404624277456647e-05, | |
| "loss": 0.3074, | |
| "step": 5495 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.9461687803268433, | |
| "eval_runtime": 53.2785, | |
| "eval_samples_per_second": 139.381, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.8689400687110319, | |
| "step": 5495 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.051445086705202e-05, | |
| "loss": 0.2883, | |
| "step": 6106 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.5767468214035034, | |
| "eval_runtime": 53.0357, | |
| "eval_samples_per_second": 140.019, | |
| "eval_steps_per_second": 4.393, | |
| "eval_wer": 0.8871357679094032, | |
| "step": 6106 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 6.698843930635838e-05, | |
| "loss": 0.2749, | |
| "step": 6716 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.3259695768356323, | |
| "eval_runtime": 53.7403, | |
| "eval_samples_per_second": 138.183, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.8212058967880319, | |
| "step": 6716 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 6.345664739884394e-05, | |
| "loss": 0.2563, | |
| "step": 7327 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.8269567489624023, | |
| "eval_runtime": 53.4539, | |
| "eval_samples_per_second": 138.923, | |
| "eval_steps_per_second": 4.359, | |
| "eval_wer": 0.844000508970607, | |
| "step": 7327 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 5.992485549132948e-05, | |
| "loss": 0.2474, | |
| "step": 7938 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 1.8218885660171509, | |
| "eval_runtime": 53.3104, | |
| "eval_samples_per_second": 139.297, | |
| "eval_steps_per_second": 4.371, | |
| "eval_wer": 0.8676676421936633, | |
| "step": 7938 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 5.6398843930635835e-05, | |
| "loss": 0.2347, | |
| "step": 8548 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.5345903635025024, | |
| "eval_runtime": 53.5574, | |
| "eval_samples_per_second": 138.655, | |
| "eval_steps_per_second": 4.35, | |
| "eval_wer": 0.8634504571646702, | |
| "step": 8548 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 5.28728323699422e-05, | |
| "loss": 0.2211, | |
| "step": 9159 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.7185391187667847, | |
| "eval_runtime": 53.0996, | |
| "eval_samples_per_second": 139.85, | |
| "eval_steps_per_second": 4.388, | |
| "eval_wer": 0.8636140548597604, | |
| "step": 9159 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 4.934104046242775e-05, | |
| "loss": 0.2117, | |
| "step": 9770 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.8662818670272827, | |
| "eval_runtime": 54.2061, | |
| "eval_samples_per_second": 136.996, | |
| "eval_steps_per_second": 4.298, | |
| "eval_wer": 0.8696671695780998, | |
| "step": 9770 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 4.582080924855491e-05, | |
| "loss": 0.1987, | |
| "step": 10380 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 1.4298427104949951, | |
| "eval_runtime": 53.5752, | |
| "eval_samples_per_second": 138.609, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.8686674058858815, | |
| "step": 10380 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 4.228901734104046e-05, | |
| "loss": 0.1814, | |
| "step": 10991 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.5630472898483276, | |
| "eval_runtime": 54.0558, | |
| "eval_samples_per_second": 137.377, | |
| "eval_steps_per_second": 4.31, | |
| "eval_wer": 0.8680130151055205, | |
| "step": 10991 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 3.8763005780346824e-05, | |
| "loss": 0.1694, | |
| "step": 11601 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 1.3626692295074463, | |
| "eval_runtime": 53.8757, | |
| "eval_samples_per_second": 137.836, | |
| "eval_steps_per_second": 4.325, | |
| "eval_wer": 0.8573246323596241, | |
| "step": 11601 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 3.523121387283237e-05, | |
| "loss": 0.1597, | |
| "step": 12212 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.7108192443847656, | |
| "eval_runtime": 54.0651, | |
| "eval_samples_per_second": 137.353, | |
| "eval_steps_per_second": 4.31, | |
| "eval_wer": 0.8641775580317379, | |
| "step": 12212 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 3.169942196531792e-05, | |
| "loss": 0.1517, | |
| "step": 12823 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 1.8344017267227173, | |
| "eval_runtime": 53.2307, | |
| "eval_samples_per_second": 139.506, | |
| "eval_steps_per_second": 4.377, | |
| "eval_wer": 0.8794466762401614, | |
| "step": 12823 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 2.817341040462428e-05, | |
| "loss": 0.1405, | |
| "step": 13433 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.4837909936904907, | |
| "eval_runtime": 52.8743, | |
| "eval_samples_per_second": 140.446, | |
| "eval_steps_per_second": 4.407, | |
| "eval_wer": 0.8508352571210441, | |
| "step": 13433 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 2.4641618497109827e-05, | |
| "loss": 0.1262, | |
| "step": 14044 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 1.5321871042251587, | |
| "eval_runtime": 57.0975, | |
| "eval_samples_per_second": 130.058, | |
| "eval_steps_per_second": 4.081, | |
| "eval_wer": 0.8415283660225765, | |
| "step": 14044 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 2.1109826589595376e-05, | |
| "loss": 0.1171, | |
| "step": 14655 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.7094522714614868, | |
| "eval_runtime": 54.2703, | |
| "eval_samples_per_second": 136.834, | |
| "eval_steps_per_second": 4.293, | |
| "eval_wer": 0.8682493228873176, | |
| "step": 14655 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 1.7589595375722546e-05, | |
| "loss": 0.1079, | |
| "step": 15265 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 1.744529128074646, | |
| "eval_runtime": 53.2445, | |
| "eval_samples_per_second": 139.47, | |
| "eval_steps_per_second": 4.376, | |
| "eval_wer": 0.8719211822660099, | |
| "step": 15265 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 1.4057803468208095e-05, | |
| "loss": 0.0996, | |
| "step": 15876 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 1.73218834400177, | |
| "eval_runtime": 55.182, | |
| "eval_samples_per_second": 134.573, | |
| "eval_steps_per_second": 4.222, | |
| "eval_wer": 0.8501808663406831, | |
| "step": 15876 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 1.0537572254335262e-05, | |
| "loss": 0.0922, | |
| "step": 16486 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 1.8349395990371704, | |
| "eval_runtime": 55.3784, | |
| "eval_samples_per_second": 134.096, | |
| "eval_steps_per_second": 4.207, | |
| "eval_wer": 0.862450693472452, | |
| "step": 16486 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 7.00578034682081e-06, | |
| "loss": 0.0855, | |
| "step": 17097 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 1.8259180784225464, | |
| "eval_runtime": 52.9961, | |
| "eval_samples_per_second": 140.124, | |
| "eval_steps_per_second": 4.397, | |
| "eval_wer": 0.8645592859869485, | |
| "step": 17097 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 3.473988439306359e-06, | |
| "loss": 0.081, | |
| "step": 17708 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 1.818723440170288, | |
| "eval_runtime": 53.212, | |
| "eval_samples_per_second": 139.555, | |
| "eval_steps_per_second": 4.379, | |
| "eval_wer": 0.8650864341155727, | |
| "step": 17708 | |
| }, | |
| { | |
| "epoch": 29.97, | |
| "learning_rate": 5.2023121387283234e-08, | |
| "loss": 0.0771, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 29.97, | |
| "eval_loss": 1.842650055885315, | |
| "eval_runtime": 54.1155, | |
| "eval_samples_per_second": 137.225, | |
| "eval_steps_per_second": 4.306, | |
| "eval_wer": 0.8623598058640685, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 29.97, | |
| "step": 18300, | |
| "total_flos": 2.0400693356246094e+21, | |
| "train_loss": 0.372250930494298, | |
| "train_runtime": 52675.4843, | |
| "train_samples_per_second": 89.021, | |
| "train_steps_per_second": 0.347 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 18300, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 2.0400693356246094e+21, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |