| { | |
| "best_metric": 0.08583583583583583, | |
| "best_model_checkpoint": "w2v2_ablation_focal_ctc_a0.75_g2.0-best_on-ling_head-tp0.025_tl10_fp0.001_fl16/checkpoint-4400", | |
| "epoch": 100.0, | |
| "eval_steps": 100, | |
| "global_step": 10600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.2830188679245284e-06, | |
| "loss": 1337.3802, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 875.6535034179688, | |
| "eval_runtime": 3.072, | |
| "eval_samples_per_second": 233.397, | |
| "eval_steps_per_second": 3.906, | |
| "eval_wer": 18.64039039039039, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3.169811320754717e-06, | |
| "loss": 928.4498, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 336.85919189453125, | |
| "eval_runtime": 2.5032, | |
| "eval_samples_per_second": 286.428, | |
| "eval_steps_per_second": 4.794, | |
| "eval_wer": 17.08541875208542, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.056603773584906e-06, | |
| "loss": 159.8141, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 65.91426086425781, | |
| "eval_runtime": 2.3194, | |
| "eval_samples_per_second": 309.133, | |
| "eval_steps_per_second": 5.174, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 6.943396226415094e-06, | |
| "loss": 84.4352, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 60.37297439575195, | |
| "eval_runtime": 2.6764, | |
| "eval_samples_per_second": 267.896, | |
| "eval_steps_per_second": 4.484, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 8.830188679245285e-06, | |
| "loss": 77.6086, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_loss": 57.35933303833008, | |
| "eval_runtime": 2.6396, | |
| "eval_samples_per_second": 271.633, | |
| "eval_steps_per_second": 4.546, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 1.0716981132075473e-05, | |
| "loss": 74.6091, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_loss": 56.16162109375, | |
| "eval_runtime": 3.0321, | |
| "eval_samples_per_second": 236.471, | |
| "eval_steps_per_second": 3.958, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 1.2603773584905661e-05, | |
| "loss": 73.5983, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "eval_loss": 55.27739334106445, | |
| "eval_runtime": 2.353, | |
| "eval_samples_per_second": 304.721, | |
| "eval_steps_per_second": 5.1, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 1.449056603773585e-05, | |
| "loss": 72.9967, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "eval_loss": 54.65113067626953, | |
| "eval_runtime": 2.238, | |
| "eval_samples_per_second": 320.375, | |
| "eval_steps_per_second": 5.362, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 1.637735849056604e-05, | |
| "loss": 71.2266, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "eval_loss": 54.53617477416992, | |
| "eval_runtime": 2.445, | |
| "eval_samples_per_second": 293.257, | |
| "eval_steps_per_second": 4.908, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 1.8264150943396227e-05, | |
| "loss": 69.7741, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_loss": 51.871822357177734, | |
| "eval_runtime": 2.3898, | |
| "eval_samples_per_second": 300.02, | |
| "eval_steps_per_second": 5.021, | |
| "eval_wer": 0.9647981314647981, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 1.999996529814242e-05, | |
| "loss": 58.1878, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "eval_loss": 28.900115966796875, | |
| "eval_runtime": 2.2862, | |
| "eval_samples_per_second": 313.622, | |
| "eval_steps_per_second": 5.249, | |
| "eval_wer": 0.5654821488154822, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 1.9993676249406895e-05, | |
| "loss": 32.9238, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "eval_loss": 12.709677696228027, | |
| "eval_runtime": 2.4457, | |
| "eval_samples_per_second": 293.166, | |
| "eval_steps_per_second": 4.907, | |
| "eval_wer": 0.2390724057390724, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 1.997655070093429e-05, | |
| "loss": 21.0735, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "eval_loss": 8.588455200195312, | |
| "eval_runtime": 2.5444, | |
| "eval_samples_per_second": 281.79, | |
| "eval_steps_per_second": 4.716, | |
| "eval_wer": 0.17851184517851185, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 1.994860722256786e-05, | |
| "loss": 15.9281, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "eval_loss": 6.895908832550049, | |
| "eval_runtime": 2.3592, | |
| "eval_samples_per_second": 303.92, | |
| "eval_steps_per_second": 5.087, | |
| "eval_wer": 0.1529029029029029, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 1.9909876114418242e-05, | |
| "loss": 13.7108, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "eval_loss": 5.751368045806885, | |
| "eval_runtime": 2.4597, | |
| "eval_samples_per_second": 291.499, | |
| "eval_steps_per_second": 4.879, | |
| "eval_wer": 0.13922255588922255, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "learning_rate": 1.9860399374007944e-05, | |
| "loss": 11.2293, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "eval_loss": 4.973855495452881, | |
| "eval_runtime": 2.554, | |
| "eval_samples_per_second": 280.739, | |
| "eval_steps_per_second": 4.699, | |
| "eval_wer": 0.12437437437437437, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 1.980023065073195e-05, | |
| "loss": 10.3682, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "eval_loss": 4.508381366729736, | |
| "eval_runtime": 2.2355, | |
| "eval_samples_per_second": 320.735, | |
| "eval_steps_per_second": 5.368, | |
| "eval_wer": 0.12370704037370704, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 1.972943518768377e-05, | |
| "loss": 9.6654, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_loss": 4.370257377624512, | |
| "eval_runtime": 2.461, | |
| "eval_samples_per_second": 291.343, | |
| "eval_steps_per_second": 4.876, | |
| "eval_wer": 0.12587587587587587, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 1.964808975090999e-05, | |
| "loss": 8.816, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_loss": 4.127786636352539, | |
| "eval_runtime": 2.5207, | |
| "eval_samples_per_second": 284.449, | |
| "eval_steps_per_second": 4.761, | |
| "eval_wer": 0.11428094761428095, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "learning_rate": 1.955628254617017e-05, | |
| "loss": 8.8608, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "eval_loss": 3.910537004470825, | |
| "eval_runtime": 2.4695, | |
| "eval_samples_per_second": 290.343, | |
| "eval_steps_per_second": 4.859, | |
| "eval_wer": 0.10744077410744077, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "learning_rate": 1.9454113123292133e-05, | |
| "loss": 7.8629, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "eval_loss": 3.9113922119140625, | |
| "eval_runtime": 2.5572, | |
| "eval_samples_per_second": 280.382, | |
| "eval_steps_per_second": 4.693, | |
| "eval_wer": 0.12370704037370704, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "learning_rate": 1.9341692268226572e-05, | |
| "loss": 7.8569, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "eval_loss": 3.7353532314300537, | |
| "eval_runtime": 2.4284, | |
| "eval_samples_per_second": 295.26, | |
| "eval_steps_per_second": 4.942, | |
| "eval_wer": 0.11211211211211211, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "learning_rate": 1.921914188291787e-05, | |
| "loss": 7.3392, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "eval_loss": 3.666839599609375, | |
| "eval_runtime": 2.409, | |
| "eval_samples_per_second": 297.636, | |
| "eval_steps_per_second": 4.981, | |
| "eval_wer": 0.10560560560560561, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 1.908659485312148e-05, | |
| "loss": 7.2164, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "eval_loss": 3.5747179985046387, | |
| "eval_runtime": 2.3787, | |
| "eval_samples_per_second": 301.422, | |
| "eval_steps_per_second": 5.045, | |
| "eval_wer": 0.11277944611277944, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "learning_rate": 1.894419490431116e-05, | |
| "loss": 7.2758, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "eval_loss": 3.4933085441589355, | |
| "eval_runtime": 2.546, | |
| "eval_samples_per_second": 281.622, | |
| "eval_steps_per_second": 4.713, | |
| "eval_wer": 0.1016016016016016, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "learning_rate": 1.8792096445832317e-05, | |
| "loss": 6.4516, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "eval_loss": 3.482123613357544, | |
| "eval_runtime": 2.3838, | |
| "eval_samples_per_second": 300.782, | |
| "eval_steps_per_second": 5.034, | |
| "eval_wer": 0.09876543209876543, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "learning_rate": 1.8630464403470435e-05, | |
| "loss": 6.45, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "eval_loss": 3.3719635009765625, | |
| "eval_runtime": 2.362, | |
| "eval_samples_per_second": 303.558, | |
| "eval_steps_per_second": 5.08, | |
| "eval_wer": 0.0995995995995996, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 1.8461229667449597e-05, | |
| "loss": 6.0068, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "eval_loss": 3.442462205886841, | |
| "eval_runtime": 2.4309, | |
| "eval_samples_per_second": 294.954, | |
| "eval_steps_per_second": 4.936, | |
| "eval_wer": 0.10443777110443778, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "learning_rate": 1.828115717353417e-05, | |
| "loss": 5.5781, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "eval_loss": 3.3221378326416016, | |
| "eval_runtime": 2.3098, | |
| "eval_samples_per_second": 310.419, | |
| "eval_steps_per_second": 5.195, | |
| "eval_wer": 0.10143476810143477, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "learning_rate": 1.809210512541925e-05, | |
| "loss": 5.5837, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "eval_loss": 3.4974489212036133, | |
| "eval_runtime": 2.4817, | |
| "eval_samples_per_second": 288.914, | |
| "eval_steps_per_second": 4.835, | |
| "eval_wer": 0.1041041041041041, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 1.7894278518986088e-05, | |
| "loss": 5.7895, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "eval_loss": 3.35359263420105, | |
| "eval_runtime": 2.5133, | |
| "eval_samples_per_second": 285.278, | |
| "eval_steps_per_second": 4.775, | |
| "eval_wer": 0.09501167834501167, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "learning_rate": 1.7687891864682602e-05, | |
| "loss": 5.6272, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "eval_loss": 3.2035768032073975, | |
| "eval_runtime": 2.4143, | |
| "eval_samples_per_second": 296.982, | |
| "eval_steps_per_second": 4.97, | |
| "eval_wer": 0.09601267934601268, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "learning_rate": 1.7473168954922044e-05, | |
| "loss": 5.594, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "eval_loss": 3.1747331619262695, | |
| "eval_runtime": 2.332, | |
| "eval_samples_per_second": 307.466, | |
| "eval_steps_per_second": 5.146, | |
| "eval_wer": 0.09125792459125792, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "learning_rate": 1.7250342621416897e-05, | |
| "loss": 4.791, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "eval_loss": 3.12252140045166, | |
| "eval_runtime": 2.3979, | |
| "eval_samples_per_second": 299.007, | |
| "eval_steps_per_second": 5.004, | |
| "eval_wer": 0.10377043710377044, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 1.7019654482711144e-05, | |
| "loss": 5.0596, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "eval_loss": 3.2113351821899414, | |
| "eval_runtime": 2.2358, | |
| "eval_samples_per_second": 320.694, | |
| "eval_steps_per_second": 5.367, | |
| "eval_wer": 0.1095261928595262, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "learning_rate": 1.6781354682184668e-05, | |
| "loss": 4.985, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "eval_loss": 3.062237024307251, | |
| "eval_runtime": 2.3928, | |
| "eval_samples_per_second": 299.643, | |
| "eval_steps_per_second": 5.015, | |
| "eval_wer": 0.09292625959292626, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "learning_rate": 1.6535701616813845e-05, | |
| "loss": 4.731, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "eval_loss": 3.0940449237823486, | |
| "eval_runtime": 2.4517, | |
| "eval_samples_per_second": 292.45, | |
| "eval_steps_per_second": 4.895, | |
| "eval_wer": 0.09559559559559559, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "learning_rate": 1.628296165698249e-05, | |
| "loss": 4.6287, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "eval_loss": 3.0453147888183594, | |
| "eval_runtime": 2.5566, | |
| "eval_samples_per_second": 280.449, | |
| "eval_steps_per_second": 4.694, | |
| "eval_wer": 0.0960960960960961, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "learning_rate": 1.6023408857646953e-05, | |
| "loss": 4.5235, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "eval_loss": 3.035074472427368, | |
| "eval_runtime": 2.4583, | |
| "eval_samples_per_second": 291.661, | |
| "eval_steps_per_second": 4.881, | |
| "eval_wer": 0.10193526860193527, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "learning_rate": 1.575732466116854e-05, | |
| "loss": 4.7715, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "eval_loss": 3.023735761642456, | |
| "eval_runtime": 2.3108, | |
| "eval_samples_per_second": 310.281, | |
| "eval_steps_per_second": 5.193, | |
| "eval_wer": 0.09275942609275943, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 1.548499759213552e-05, | |
| "loss": 4.7101, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "eval_loss": 3.024998426437378, | |
| "eval_runtime": 2.2347, | |
| "eval_samples_per_second": 320.853, | |
| "eval_steps_per_second": 5.37, | |
| "eval_wer": 0.09426092759426093, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "learning_rate": 1.5206722944505621e-05, | |
| "loss": 4.243, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "eval_loss": 2.9704301357269287, | |
| "eval_runtime": 2.278, | |
| "eval_samples_per_second": 314.754, | |
| "eval_steps_per_second": 5.268, | |
| "eval_wer": 0.09801468134801468, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "learning_rate": 1.4922802461408265e-05, | |
| "loss": 4.4015, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "eval_loss": 2.9600486755371094, | |
| "eval_runtime": 2.138, | |
| "eval_samples_per_second": 335.367, | |
| "eval_steps_per_second": 5.613, | |
| "eval_wer": 0.08708708708708708, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "learning_rate": 1.4633544007953716e-05, | |
| "loss": 4.4545, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "eval_loss": 2.980609655380249, | |
| "eval_runtime": 2.3548, | |
| "eval_samples_per_second": 304.478, | |
| "eval_steps_per_second": 5.096, | |
| "eval_wer": 0.08583583583583583, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "learning_rate": 1.4339261237403948e-05, | |
| "loss": 4.662, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "eval_loss": 2.9668173789978027, | |
| "eval_runtime": 2.2963, | |
| "eval_samples_per_second": 312.242, | |
| "eval_steps_per_second": 5.226, | |
| "eval_wer": 0.09693026359693026, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "learning_rate": 1.4043285361283684e-05, | |
| "loss": 4.0696, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "eval_loss": 2.9349417686462402, | |
| "eval_runtime": 2.305, | |
| "eval_samples_per_second": 311.064, | |
| "eval_steps_per_second": 5.206, | |
| "eval_wer": 0.09351017684351018, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "learning_rate": 1.3739958551927287e-05, | |
| "loss": 3.5668, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "eval_loss": 2.9190385341644287, | |
| "eval_runtime": 2.3277, | |
| "eval_samples_per_second": 308.023, | |
| "eval_steps_per_second": 5.155, | |
| "eval_wer": 0.09167500834167501, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "learning_rate": 1.3432576372069266e-05, | |
| "loss": 3.8214, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "eval_loss": 2.948979139328003, | |
| "eval_runtime": 2.1809, | |
| "eval_samples_per_second": 328.766, | |
| "eval_steps_per_second": 5.502, | |
| "eval_wer": 0.09009009009009009, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "learning_rate": 1.3121472127174802e-05, | |
| "loss": 3.8215, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "eval_loss": 2.937129497528076, | |
| "eval_runtime": 2.441, | |
| "eval_samples_per_second": 293.729, | |
| "eval_steps_per_second": 4.916, | |
| "eval_wer": 0.0911745078411745, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "learning_rate": 1.2806983158677e-05, | |
| "loss": 3.6593, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "eval_loss": 2.9407706260681152, | |
| "eval_runtime": 2.3032, | |
| "eval_samples_per_second": 311.309, | |
| "eval_steps_per_second": 5.21, | |
| "eval_wer": 0.08750417083750417, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "learning_rate": 1.248945047818555e-05, | |
| "loss": 3.3709, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "eval_loss": 2.9576809406280518, | |
| "eval_runtime": 2.3035, | |
| "eval_samples_per_second": 311.269, | |
| "eval_steps_per_second": 5.21, | |
| "eval_wer": 0.09200867534200867, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "learning_rate": 1.2169218397715645e-05, | |
| "loss": 3.5768, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "eval_loss": 2.986335039138794, | |
| "eval_runtime": 2.3677, | |
| "eval_samples_per_second": 302.826, | |
| "eval_steps_per_second": 5.068, | |
| "eval_wer": 0.09401067734401068, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.1846634156338193e-05, | |
| "loss": 3.3018, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 2.943732500076294, | |
| "eval_runtime": 2.3937, | |
| "eval_samples_per_second": 299.537, | |
| "eval_steps_per_second": 5.013, | |
| "eval_wer": 0.10026693360026694, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 50.94, | |
| "learning_rate": 1.1522047543656106e-05, | |
| "loss": 3.2921, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 50.94, | |
| "eval_loss": 2.9195058345794678, | |
| "eval_runtime": 2.5086, | |
| "eval_samples_per_second": 285.812, | |
| "eval_steps_per_second": 4.783, | |
| "eval_wer": 0.09225892559225893, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "learning_rate": 1.119581052051494e-05, | |
| "loss": 3.4551, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "eval_loss": 2.940958261489868, | |
| "eval_runtime": 2.5071, | |
| "eval_samples_per_second": 285.992, | |
| "eval_steps_per_second": 4.786, | |
| "eval_wer": 0.09501167834501167, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 52.83, | |
| "learning_rate": 1.0868276837359204e-05, | |
| "loss": 3.6576, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 52.83, | |
| "eval_loss": 2.9519896507263184, | |
| "eval_runtime": 2.1953, | |
| "eval_samples_per_second": 326.602, | |
| "eval_steps_per_second": 5.466, | |
| "eval_wer": 0.1011011011011011, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 53.77, | |
| "learning_rate": 1.0539801650648128e-05, | |
| "loss": 3.5078, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 53.77, | |
| "eval_loss": 2.8925814628601074, | |
| "eval_runtime": 2.2343, | |
| "eval_samples_per_second": 320.899, | |
| "eval_steps_per_second": 5.371, | |
| "eval_wer": 0.09367701034367701, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 54.72, | |
| "learning_rate": 1.0210741137746833e-05, | |
| "loss": 3.0777, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 54.72, | |
| "eval_loss": 2.8970932960510254, | |
| "eval_runtime": 2.2598, | |
| "eval_samples_per_second": 317.283, | |
| "eval_steps_per_second": 5.31, | |
| "eval_wer": 0.09134134134134134, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 55.66, | |
| "learning_rate": 9.881452110710519e-06, | |
| "loss": 3.0572, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 55.66, | |
| "eval_loss": 2.8692612648010254, | |
| "eval_runtime": 2.2528, | |
| "eval_samples_per_second": 318.271, | |
| "eval_steps_per_second": 5.327, | |
| "eval_wer": 0.0890890890890891, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 56.6, | |
| "learning_rate": 9.552291629380419e-06, | |
| "loss": 3.0486, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 56.6, | |
| "eval_loss": 2.8876469135284424, | |
| "eval_runtime": 2.2032, | |
| "eval_samples_per_second": 325.435, | |
| "eval_steps_per_second": 5.447, | |
| "eval_wer": 0.0881715048381715, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 57.55, | |
| "learning_rate": 9.223616614211057e-06, | |
| "loss": 3.1283, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 57.55, | |
| "eval_loss": 2.8596761226654053, | |
| "eval_runtime": 2.2244, | |
| "eval_samples_per_second": 322.333, | |
| "eval_steps_per_second": 5.395, | |
| "eval_wer": 0.09134134134134134, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 58.49, | |
| "learning_rate": 8.895783459248695e-06, | |
| "loss": 2.8705, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 58.49, | |
| "eval_loss": 2.9079818725585938, | |
| "eval_runtime": 2.3678, | |
| "eval_samples_per_second": 302.819, | |
| "eval_steps_per_second": 5.068, | |
| "eval_wer": 0.09042375709042376, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 59.43, | |
| "learning_rate": 8.569147645680523e-06, | |
| "loss": 3.0644, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 59.43, | |
| "eval_loss": 2.910602331161499, | |
| "eval_runtime": 2.3829, | |
| "eval_samples_per_second": 300.896, | |
| "eval_steps_per_second": 5.036, | |
| "eval_wer": 0.09167500834167501, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 60.38, | |
| "learning_rate": 8.244063356373729e-06, | |
| "loss": 2.8822, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 60.38, | |
| "eval_loss": 2.9231059551239014, | |
| "eval_runtime": 2.4556, | |
| "eval_samples_per_second": 291.984, | |
| "eval_steps_per_second": 4.887, | |
| "eval_wer": 0.0890890890890891, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 61.32, | |
| "learning_rate": 7.92088309182241e-06, | |
| "loss": 3.2338, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 61.32, | |
| "eval_loss": 2.9510579109191895, | |
| "eval_runtime": 2.1932, | |
| "eval_samples_per_second": 326.917, | |
| "eval_steps_per_second": 5.471, | |
| "eval_wer": 0.09034034034034034, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 62.26, | |
| "learning_rate": 7.599957287918655e-06, | |
| "loss": 3.048, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 62.26, | |
| "eval_loss": 2.953885316848755, | |
| "eval_runtime": 2.3592, | |
| "eval_samples_per_second": 303.918, | |
| "eval_steps_per_second": 5.086, | |
| "eval_wer": 0.08975642308975643, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 63.21, | |
| "learning_rate": 7.281633935962448e-06, | |
| "loss": 3.094, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 63.21, | |
| "eval_loss": 2.949049949645996, | |
| "eval_runtime": 2.5261, | |
| "eval_samples_per_second": 283.836, | |
| "eval_steps_per_second": 4.75, | |
| "eval_wer": 0.09084084084084085, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 64.15, | |
| "learning_rate": 6.966258205322274e-06, | |
| "loss": 3.0581, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 64.15, | |
| "eval_loss": 2.895172119140625, | |
| "eval_runtime": 1.5993, | |
| "eval_samples_per_second": 448.308, | |
| "eval_steps_per_second": 7.503, | |
| "eval_wer": 0.08858858858858859, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 65.09, | |
| "learning_rate": 6.654172069155635e-06, | |
| "loss": 2.9343, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 65.09, | |
| "eval_loss": 2.8925557136535645, | |
| "eval_runtime": 2.4073, | |
| "eval_samples_per_second": 297.843, | |
| "eval_steps_per_second": 4.985, | |
| "eval_wer": 0.08825492158825492, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "learning_rate": 6.34571393359539e-06, | |
| "loss": 2.9497, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "eval_loss": 2.8731818199157715, | |
| "eval_runtime": 2.4257, | |
| "eval_samples_per_second": 295.589, | |
| "eval_steps_per_second": 4.947, | |
| "eval_wer": 0.08875542208875542, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 66.98, | |
| "learning_rate": 6.0412182708038794e-06, | |
| "loss": 2.7788, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 66.98, | |
| "eval_loss": 2.8836898803710938, | |
| "eval_runtime": 2.3112, | |
| "eval_samples_per_second": 310.231, | |
| "eval_steps_per_second": 5.192, | |
| "eval_wer": 0.09042375709042376, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 67.92, | |
| "learning_rate": 5.741015256292837e-06, | |
| "loss": 2.7765, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 67.92, | |
| "eval_loss": 2.9168927669525146, | |
| "eval_runtime": 2.2787, | |
| "eval_samples_per_second": 314.657, | |
| "eval_steps_per_second": 5.266, | |
| "eval_wer": 0.09509509509509509, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 68.87, | |
| "learning_rate": 5.445430410902306e-06, | |
| "loss": 3.134, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 68.87, | |
| "eval_loss": 2.903002977371216, | |
| "eval_runtime": 2.3536, | |
| "eval_samples_per_second": 304.633, | |
| "eval_steps_per_second": 5.098, | |
| "eval_wer": 0.09259259259259259, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 69.81, | |
| "learning_rate": 5.15478424782678e-06, | |
| "loss": 2.8812, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 69.81, | |
| "eval_loss": 2.9044971466064453, | |
| "eval_runtime": 2.2556, | |
| "eval_samples_per_second": 317.872, | |
| "eval_steps_per_second": 5.32, | |
| "eval_wer": 0.0920920920920921, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "learning_rate": 4.869391925071285e-06, | |
| "loss": 2.615, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "eval_loss": 2.9148054122924805, | |
| "eval_runtime": 2.2984, | |
| "eval_samples_per_second": 311.958, | |
| "eval_steps_per_second": 5.221, | |
| "eval_wer": 0.08708708708708708, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 71.7, | |
| "learning_rate": 4.589562903714348e-06, | |
| "loss": 2.5678, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 71.7, | |
| "eval_loss": 2.943516969680786, | |
| "eval_runtime": 2.264, | |
| "eval_samples_per_second": 316.693, | |
| "eval_steps_per_second": 5.3, | |
| "eval_wer": 0.09217550884217551, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 72.64, | |
| "learning_rate": 4.318310214093595e-06, | |
| "loss": 2.4858, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 72.64, | |
| "eval_loss": 2.9050374031066895, | |
| "eval_runtime": 2.5001, | |
| "eval_samples_per_second": 286.789, | |
| "eval_steps_per_second": 4.8, | |
| "eval_wer": 0.09284284284284285, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 73.58, | |
| "learning_rate": 4.050448633035326e-06, | |
| "loss": 2.5367, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 73.58, | |
| "eval_loss": 2.8948252201080322, | |
| "eval_runtime": 2.3819, | |
| "eval_samples_per_second": 301.016, | |
| "eval_steps_per_second": 5.038, | |
| "eval_wer": 0.08775442108775443, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "learning_rate": 3.7890383628075156e-06, | |
| "loss": 2.3228, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "eval_loss": 2.8994953632354736, | |
| "eval_runtime": 2.4234, | |
| "eval_samples_per_second": 295.866, | |
| "eval_steps_per_second": 4.952, | |
| "eval_wer": 0.0890890890890891, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 75.47, | |
| "learning_rate": 3.5343628598934275e-06, | |
| "loss": 2.5849, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 75.47, | |
| "eval_loss": 2.9289326667785645, | |
| "eval_runtime": 2.4566, | |
| "eval_samples_per_second": 291.864, | |
| "eval_steps_per_second": 4.885, | |
| "eval_wer": 0.09275942609275943, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 76.42, | |
| "learning_rate": 3.2866982780278357e-06, | |
| "loss": 2.6645, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 76.42, | |
| "eval_loss": 2.8950443267822266, | |
| "eval_runtime": 2.3369, | |
| "eval_samples_per_second": 306.813, | |
| "eval_steps_per_second": 5.135, | |
| "eval_wer": 0.08842175508842176, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 77.36, | |
| "learning_rate": 3.0463131687536695e-06, | |
| "loss": 2.6634, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 77.36, | |
| "eval_loss": 2.919368267059326, | |
| "eval_runtime": 2.2121, | |
| "eval_samples_per_second": 324.122, | |
| "eval_steps_per_second": 5.425, | |
| "eval_wer": 0.09217550884217551, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 78.3, | |
| "learning_rate": 2.8134681902219797e-06, | |
| "loss": 2.393, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 78.3, | |
| "eval_loss": 2.9073572158813477, | |
| "eval_runtime": 2.3983, | |
| "eval_samples_per_second": 298.96, | |
| "eval_steps_per_second": 5.004, | |
| "eval_wer": 0.09192525859192525, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 79.25, | |
| "learning_rate": 2.5884158245510325e-06, | |
| "loss": 3.0675, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 79.25, | |
| "eval_loss": 2.892658233642578, | |
| "eval_runtime": 1.8302, | |
| "eval_samples_per_second": 391.764, | |
| "eval_steps_per_second": 6.557, | |
| "eval_wer": 0.09075742409075742, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 80.19, | |
| "learning_rate": 2.3714001040509095e-06, | |
| "loss": 2.6344, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 80.19, | |
| "eval_loss": 2.876831293106079, | |
| "eval_runtime": 1.6456, | |
| "eval_samples_per_second": 435.704, | |
| "eval_steps_per_second": 7.292, | |
| "eval_wer": 0.0890890890890891, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 81.13, | |
| "learning_rate": 2.162656346610592e-06, | |
| "loss": 2.5742, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 81.13, | |
| "eval_loss": 2.8808791637420654, | |
| "eval_runtime": 2.9484, | |
| "eval_samples_per_second": 243.18, | |
| "eval_steps_per_second": 4.07, | |
| "eval_wer": 0.09109109109109109, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 82.08, | |
| "learning_rate": 1.962410900534367e-06, | |
| "loss": 2.6523, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 82.08, | |
| "eval_loss": 2.863851547241211, | |
| "eval_runtime": 2.6818, | |
| "eval_samples_per_second": 267.357, | |
| "eval_steps_per_second": 4.475, | |
| "eval_wer": 0.08625291958625292, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 83.02, | |
| "learning_rate": 1.7746247286710604e-06, | |
| "loss": 2.2657, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 83.02, | |
| "eval_loss": 2.8809173107147217, | |
| "eval_runtime": 2.6706, | |
| "eval_samples_per_second": 268.481, | |
| "eval_steps_per_second": 4.493, | |
| "eval_wer": 0.0911745078411745, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 83.96, | |
| "learning_rate": 1.591837419081773e-06, | |
| "loss": 2.3238, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 83.96, | |
| "eval_loss": 2.876391887664795, | |
| "eval_runtime": 2.5394, | |
| "eval_samples_per_second": 282.355, | |
| "eval_steps_per_second": 4.726, | |
| "eval_wer": 0.08933933933933934, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 84.91, | |
| "learning_rate": 1.4181673801868301e-06, | |
| "loss": 2.3664, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 84.91, | |
| "eval_loss": 2.873842239379883, | |
| "eval_runtime": 2.4749, | |
| "eval_samples_per_second": 289.712, | |
| "eval_steps_per_second": 4.849, | |
| "eval_wer": 0.09125792459125792, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 85.85, | |
| "learning_rate": 1.2538029286060428e-06, | |
| "loss": 2.5655, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 85.85, | |
| "eval_loss": 2.887613296508789, | |
| "eval_runtime": 2.3857, | |
| "eval_samples_per_second": 300.547, | |
| "eval_steps_per_second": 5.03, | |
| "eval_wer": 0.09042375709042376, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 86.79, | |
| "learning_rate": 1.0989222905788767e-06, | |
| "loss": 2.4372, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 86.79, | |
| "eval_loss": 2.902373790740967, | |
| "eval_runtime": 2.3648, | |
| "eval_samples_per_second": 303.2, | |
| "eval_steps_per_second": 5.074, | |
| "eval_wer": 0.09100767434100768, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 87.74, | |
| "learning_rate": 9.536934087073702e-07, | |
| "loss": 2.5267, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 87.74, | |
| "eval_loss": 2.892174243927002, | |
| "eval_runtime": 2.5482, | |
| "eval_samples_per_second": 281.372, | |
| "eval_steps_per_second": 4.709, | |
| "eval_wer": 0.08975642308975643, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 88.68, | |
| "learning_rate": 8.182737598499846e-07, | |
| "loss": 2.471, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 88.68, | |
| "eval_loss": 2.889312744140625, | |
| "eval_runtime": 2.3375, | |
| "eval_samples_per_second": 306.735, | |
| "eval_steps_per_second": 5.134, | |
| "eval_wer": 0.08842175508842176, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "learning_rate": 6.928101843638202e-07, | |
| "loss": 2.5225, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "eval_loss": 2.885225296020508, | |
| "eval_runtime": 2.3766, | |
| "eval_samples_per_second": 301.687, | |
| "eval_steps_per_second": 5.049, | |
| "eval_wer": 0.08875542208875542, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 90.57, | |
| "learning_rate": 5.774387268803871e-07, | |
| "loss": 2.4752, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 90.57, | |
| "eval_loss": 2.8876028060913086, | |
| "eval_runtime": 2.4035, | |
| "eval_samples_per_second": 298.319, | |
| "eval_steps_per_second": 4.993, | |
| "eval_wer": 0.08917250583917251, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 91.51, | |
| "learning_rate": 4.722844887875522e-07, | |
| "loss": 2.5029, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 91.51, | |
| "eval_loss": 2.8883309364318848, | |
| "eval_runtime": 2.5391, | |
| "eval_samples_per_second": 282.383, | |
| "eval_steps_per_second": 4.726, | |
| "eval_wer": 0.08850517183850518, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 92.45, | |
| "learning_rate": 3.7746149257763984e-07, | |
| "loss": 2.7052, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 92.45, | |
| "eval_loss": 2.882530450820923, | |
| "eval_runtime": 2.5114, | |
| "eval_samples_per_second": 285.5, | |
| "eval_steps_per_second": 4.778, | |
| "eval_wer": 0.08708708708708708, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 93.4, | |
| "learning_rate": 2.9307255820877676e-07, | |
| "loss": 2.4682, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 93.4, | |
| "eval_loss": 2.8780291080474854, | |
| "eval_runtime": 2.3878, | |
| "eval_samples_per_second": 300.275, | |
| "eval_steps_per_second": 5.026, | |
| "eval_wer": 0.08700367033700367, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 94.34, | |
| "learning_rate": 2.1920919161354304e-07, | |
| "loss": 2.3672, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 94.34, | |
| "eval_loss": 2.8809967041015625, | |
| "eval_runtime": 2.5256, | |
| "eval_samples_per_second": 283.891, | |
| "eval_steps_per_second": 4.751, | |
| "eval_wer": 0.08717050383717051, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 95.28, | |
| "learning_rate": 1.5595148547582373e-07, | |
| "loss": 2.5325, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 95.28, | |
| "eval_loss": 2.88421630859375, | |
| "eval_runtime": 2.5311, | |
| "eval_samples_per_second": 283.279, | |
| "eval_steps_per_second": 4.741, | |
| "eval_wer": 0.08842175508842176, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 96.23, | |
| "learning_rate": 1.0336803238345072e-07, | |
| "loss": 2.4877, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 96.23, | |
| "eval_loss": 2.88330340385437, | |
| "eval_runtime": 2.3205, | |
| "eval_samples_per_second": 308.982, | |
| "eval_steps_per_second": 5.171, | |
| "eval_wer": 0.08842175508842176, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 97.17, | |
| "learning_rate": 6.151585045082286e-08, | |
| "loss": 2.7373, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 97.17, | |
| "eval_loss": 2.8824727535247803, | |
| "eval_runtime": 2.4169, | |
| "eval_samples_per_second": 296.665, | |
| "eval_steps_per_second": 4.965, | |
| "eval_wer": 0.0881715048381715, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 98.11, | |
| "learning_rate": 3.044032149211562e-08, | |
| "loss": 2.5574, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 98.11, | |
| "eval_loss": 2.8832902908325195, | |
| "eval_runtime": 2.3516, | |
| "eval_samples_per_second": 304.903, | |
| "eval_steps_per_second": 5.103, | |
| "eval_wer": 0.08808808808808809, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 99.06, | |
| "learning_rate": 1.0175141812170941e-08, | |
| "loss": 2.2097, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 99.06, | |
| "eval_loss": 2.8823037147521973, | |
| "eval_runtime": 2.2498, | |
| "eval_samples_per_second": 318.7, | |
| "eval_steps_per_second": 5.334, | |
| "eval_wer": 0.08825492158825492, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 7.422856683902968e-10, | |
| "loss": 2.5919, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 2.8828866481781006, | |
| "eval_runtime": 2.2981, | |
| "eval_samples_per_second": 312.003, | |
| "eval_steps_per_second": 5.222, | |
| "eval_wer": 0.08792125458792126, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 10600, | |
| "total_flos": 2.842046364754798e+19, | |
| "train_loss": 32.606080211423475, | |
| "train_runtime": 7534.6733, | |
| "train_samples_per_second": 44.912, | |
| "train_steps_per_second": 1.407 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 10600, | |
| "num_train_epochs": 100, | |
| "save_steps": 100, | |
| "total_flos": 2.842046364754798e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |