| { | |
| "best_metric": 0.8215018908698001, | |
| "best_model_checkpoint": "/content/wav2vec2-large-xlsr-bn/checkpoint-800", | |
| "epoch": 30.0, | |
| "global_step": 18750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00029405766150560596, | |
| "loss": 4.1339, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 3.2776501178741455, | |
| "eval_runtime": 74.0715, | |
| "eval_samples_per_second": 13.5, | |
| "eval_steps_per_second": 1.688, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00028774693005872927, | |
| "loss": 2.0718, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.979073703289032, | |
| "eval_runtime": 74.5153, | |
| "eval_samples_per_second": 13.42, | |
| "eval_steps_per_second": 1.678, | |
| "eval_wer": 0.8215018908698001, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00028143619861185264, | |
| "loss": 2.4619, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 3.080274820327759, | |
| "eval_runtime": 74.0712, | |
| "eval_samples_per_second": 13.501, | |
| "eval_steps_per_second": 1.688, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.000275157501334757, | |
| "loss": 3.1703, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 3.4948482513427734, | |
| "eval_runtime": 74.092, | |
| "eval_samples_per_second": 13.497, | |
| "eval_steps_per_second": 1.687, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00026886278697277094, | |
| "loss": 5.5212, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 74.4322, | |
| "eval_samples_per_second": 13.435, | |
| "eval_steps_per_second": 1.679, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.0002625680726107848, | |
| "loss": 5.8864, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 74.0924, | |
| "eval_samples_per_second": 13.497, | |
| "eval_steps_per_second": 1.687, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.00025628937533368924, | |
| "loss": 5.9244, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 71.9212, | |
| "eval_samples_per_second": 13.904, | |
| "eval_steps_per_second": 1.738, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 0.00024996262680192206, | |
| "loss": 5.8882, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.014, | |
| "eval_samples_per_second": 13.886, | |
| "eval_steps_per_second": 1.736, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.00024365189535504537, | |
| "loss": 5.9011, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.5441, | |
| "eval_samples_per_second": 13.785, | |
| "eval_steps_per_second": 1.723, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00023734116390816868, | |
| "loss": 5.8956, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.2548, | |
| "eval_samples_per_second": 13.84, | |
| "eval_steps_per_second": 1.73, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.00023103043246129203, | |
| "loss": 5.9103, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.8707, | |
| "eval_samples_per_second": 13.723, | |
| "eval_steps_per_second": 1.715, | |
| "eval_wer": 1.0, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 0.0002247357180993059, | |
| "loss": 5.9083, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.091, | |
| "eval_samples_per_second": 13.871, | |
| "eval_steps_per_second": 1.734, | |
| "eval_wer": 1.0, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 0.0002184410037373198, | |
| "loss": 5.9054, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 74.6941, | |
| "eval_samples_per_second": 13.388, | |
| "eval_steps_per_second": 1.673, | |
| "eval_wer": 1.0, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 0.00021213027229044314, | |
| "loss": 5.8996, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.5093, | |
| "eval_samples_per_second": 13.791, | |
| "eval_steps_per_second": 1.724, | |
| "eval_wer": 1.0, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.00020581954084356645, | |
| "loss": 5.8929, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 73.1997, | |
| "eval_samples_per_second": 13.661, | |
| "eval_steps_per_second": 1.708, | |
| "eval_wer": 1.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 0.0001995408435664709, | |
| "loss": 5.9021, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.6358, | |
| "eval_samples_per_second": 13.767, | |
| "eval_steps_per_second": 1.721, | |
| "eval_wer": 1.0, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 0.00019323011211959423, | |
| "loss": 5.8853, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.5502, | |
| "eval_samples_per_second": 13.784, | |
| "eval_steps_per_second": 1.723, | |
| "eval_wer": 1.0, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 0.000186903363587827, | |
| "loss": 5.8968, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.2966, | |
| "eval_samples_per_second": 13.832, | |
| "eval_steps_per_second": 1.729, | |
| "eval_wer": 1.0, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.00018060864922584088, | |
| "loss": 5.9139, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.4188, | |
| "eval_samples_per_second": 13.809, | |
| "eval_steps_per_second": 1.726, | |
| "eval_wer": 1.0, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.00017431393486385476, | |
| "loss": 5.8976, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.6699, | |
| "eval_samples_per_second": 13.761, | |
| "eval_steps_per_second": 1.72, | |
| "eval_wer": 1.0, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 0.00016798718633208754, | |
| "loss": 5.8909, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "eval_loss": 6.071021556854248, | |
| "eval_runtime": 72.1967, | |
| "eval_samples_per_second": 13.851, | |
| "eval_steps_per_second": 1.731, | |
| "eval_wer": 1.0, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "learning_rate": 0.00016202883075280296, | |
| "loss": 7.2473, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "eval_loss": NaN, | |
| "eval_runtime": 74.9323, | |
| "eval_samples_per_second": 13.345, | |
| "eval_steps_per_second": 1.668, | |
| "eval_wer": 1.0, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 0.000155621996796583, | |
| "loss": 0.0, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "eval_loss": NaN, | |
| "eval_runtime": 72.0418, | |
| "eval_samples_per_second": 13.881, | |
| "eval_steps_per_second": 1.735, | |
| "eval_wer": 1.0, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 15.36, | |
| "learning_rate": 0.00014921516284036303, | |
| "loss": 0.0, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 15.36, | |
| "eval_loss": NaN, | |
| "eval_runtime": 72.991, | |
| "eval_samples_per_second": 13.7, | |
| "eval_steps_per_second": 1.713, | |
| "eval_wer": 1.0, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.00014280832888414308, | |
| "loss": 0.0, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 74.3256, | |
| "eval_samples_per_second": 13.454, | |
| "eval_steps_per_second": 1.682, | |
| "eval_wer": 1.0, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 16.64, | |
| "learning_rate": 0.0001364014949279231, | |
| "loss": 0.0, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 16.64, | |
| "eval_loss": NaN, | |
| "eval_runtime": 85.1556, | |
| "eval_samples_per_second": 11.743, | |
| "eval_steps_per_second": 1.468, | |
| "eval_wer": 1.0, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 17.28, | |
| "learning_rate": 0.00012999466097170315, | |
| "loss": 0.0, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 17.28, | |
| "eval_loss": NaN, | |
| "eval_runtime": 74.297, | |
| "eval_samples_per_second": 13.46, | |
| "eval_steps_per_second": 1.682, | |
| "eval_wer": 1.0, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 0.00012358782701548318, | |
| "loss": 0.0, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_loss": NaN, | |
| "eval_runtime": 71.8705, | |
| "eval_samples_per_second": 13.914, | |
| "eval_steps_per_second": 1.739, | |
| "eval_wer": 1.0, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 0.00011718099305926321, | |
| "loss": 0.0, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "eval_loss": NaN, | |
| "eval_runtime": 72.0682, | |
| "eval_samples_per_second": 13.876, | |
| "eval_steps_per_second": 1.734, | |
| "eval_wer": 1.0, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 0.00011077415910304323, | |
| "loss": 0.0, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "eval_loss": NaN, | |
| "eval_runtime": 86.6422, | |
| "eval_samples_per_second": 11.542, | |
| "eval_steps_per_second": 1.443, | |
| "eval_wer": 1.0, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "learning_rate": 0.00010436732514682327, | |
| "loss": 0.0, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 19.84, | |
| "eval_loss": NaN, | |
| "eval_runtime": 74.8452, | |
| "eval_samples_per_second": 13.361, | |
| "eval_steps_per_second": 1.67, | |
| "eval_wer": 1.0, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 20.48, | |
| "learning_rate": 9.79604911906033e-05, | |
| "loss": 0.0, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 20.48, | |
| "eval_loss": NaN, | |
| "eval_runtime": 73.8255, | |
| "eval_samples_per_second": 13.545, | |
| "eval_steps_per_second": 1.693, | |
| "eval_wer": 1.0, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 21.12, | |
| "learning_rate": 9.155365723438334e-05, | |
| "loss": 0.0, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 21.12, | |
| "eval_loss": NaN, | |
| "eval_runtime": 77.5777, | |
| "eval_samples_per_second": 12.89, | |
| "eval_steps_per_second": 1.611, | |
| "eval_wer": 1.0, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 21.76, | |
| "learning_rate": 8.514682327816338e-05, | |
| "loss": 0.0, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 21.76, | |
| "eval_loss": NaN, | |
| "eval_runtime": 78.2086, | |
| "eval_samples_per_second": 12.786, | |
| "eval_steps_per_second": 1.598, | |
| "eval_wer": 1.0, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 7.87399893219434e-05, | |
| "loss": 0.0, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "eval_loss": NaN, | |
| "eval_runtime": 81.9911, | |
| "eval_samples_per_second": 12.196, | |
| "eval_steps_per_second": 1.525, | |
| "eval_wer": 1.0, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 23.04, | |
| "learning_rate": 7.233315536572343e-05, | |
| "loss": 0.0, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 23.04, | |
| "eval_loss": NaN, | |
| "eval_runtime": 74.0428, | |
| "eval_samples_per_second": 13.506, | |
| "eval_steps_per_second": 1.688, | |
| "eval_wer": 1.0, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 6.592632140950347e-05, | |
| "loss": 0.0, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "eval_loss": NaN, | |
| "eval_runtime": 73.6487, | |
| "eval_samples_per_second": 13.578, | |
| "eval_steps_per_second": 1.697, | |
| "eval_wer": 1.0, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 24.32, | |
| "learning_rate": 5.95194874532835e-05, | |
| "loss": 0.0, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 24.32, | |
| "eval_loss": NaN, | |
| "eval_runtime": 73.0072, | |
| "eval_samples_per_second": 13.697, | |
| "eval_steps_per_second": 1.712, | |
| "eval_wer": 1.0, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "learning_rate": 5.3112653497063527e-05, | |
| "loss": 0.0, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "eval_loss": NaN, | |
| "eval_runtime": 79.0121, | |
| "eval_samples_per_second": 12.656, | |
| "eval_steps_per_second": 1.582, | |
| "eval_wer": 1.0, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "learning_rate": 4.670581954084356e-05, | |
| "loss": 0.0, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "eval_loss": NaN, | |
| "eval_runtime": 82.7961, | |
| "eval_samples_per_second": 12.078, | |
| "eval_steps_per_second": 1.51, | |
| "eval_wer": 1.0, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 4.02989855846236e-05, | |
| "loss": 0.0, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "eval_loss": NaN, | |
| "eval_runtime": 83.3819, | |
| "eval_samples_per_second": 11.993, | |
| "eval_steps_per_second": 1.499, | |
| "eval_wer": 1.0, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "learning_rate": 3.3892151628403626e-05, | |
| "loss": 0.0, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "eval_loss": NaN, | |
| "eval_runtime": 79.5109, | |
| "eval_samples_per_second": 12.577, | |
| "eval_steps_per_second": 1.572, | |
| "eval_wer": 1.0, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 27.52, | |
| "learning_rate": 2.748531767218366e-05, | |
| "loss": 0.0, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 27.52, | |
| "eval_loss": NaN, | |
| "eval_runtime": 76.3506, | |
| "eval_samples_per_second": 13.097, | |
| "eval_steps_per_second": 1.637, | |
| "eval_wer": 1.0, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 28.16, | |
| "learning_rate": 2.107848371596369e-05, | |
| "loss": 0.0, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 28.16, | |
| "eval_loss": NaN, | |
| "eval_runtime": 72.6042, | |
| "eval_samples_per_second": 13.773, | |
| "eval_steps_per_second": 1.722, | |
| "eval_wer": 1.0, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "learning_rate": 1.4671649759743725e-05, | |
| "loss": 0.0, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "eval_loss": NaN, | |
| "eval_runtime": 83.2188, | |
| "eval_samples_per_second": 12.017, | |
| "eval_steps_per_second": 1.502, | |
| "eval_wer": 1.0, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 29.44, | |
| "learning_rate": 8.264815803523758e-06, | |
| "loss": 0.0, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 29.44, | |
| "eval_loss": NaN, | |
| "eval_runtime": 76.8024, | |
| "eval_samples_per_second": 13.02, | |
| "eval_steps_per_second": 1.628, | |
| "eval_wer": 1.0, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 18750, | |
| "total_flos": 1.1136444297962675e+20, | |
| "train_loss": 2.53877658203125, | |
| "train_runtime": 56941.5798, | |
| "train_samples_per_second": 10.537, | |
| "train_steps_per_second": 0.329 | |
| } | |
| ], | |
| "max_steps": 18750, | |
| "num_train_epochs": 30, | |
| "total_flos": 1.1136444297962675e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |