| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9357336430507162, | |
| "eval_steps": 100, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019357336430507164, | |
| "eval_loss": 3.554515838623047, | |
| "eval_runtime": 148.6327, | |
| "eval_samples_per_second": 38.054, | |
| "eval_steps_per_second": 4.757, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03871467286101433, | |
| "eval_loss": 3.02604603767395, | |
| "eval_runtime": 146.1639, | |
| "eval_samples_per_second": 38.696, | |
| "eval_steps_per_second": 4.837, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05807200929152149, | |
| "eval_loss": 2.906620979309082, | |
| "eval_runtime": 146.0171, | |
| "eval_samples_per_second": 38.735, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07742934572202866, | |
| "eval_loss": 2.013317823410034, | |
| "eval_runtime": 145.6964, | |
| "eval_samples_per_second": 38.82, | |
| "eval_steps_per_second": 4.853, | |
| "eval_wer": 0.9847217987193272, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "grad_norm": 4.750732421875, | |
| "learning_rate": 0.0002982, | |
| "loss": 4.0489, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "eval_loss": 1.459844708442688, | |
| "eval_runtime": 145.9454, | |
| "eval_samples_per_second": 38.754, | |
| "eval_steps_per_second": 4.844, | |
| "eval_wer": 0.9003546725297299, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11614401858304298, | |
| "eval_loss": 1.1772356033325195, | |
| "eval_runtime": 145.9941, | |
| "eval_samples_per_second": 38.741, | |
| "eval_steps_per_second": 4.843, | |
| "eval_wer": 0.8042079247644878, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13550135501355012, | |
| "eval_loss": 1.07865571975708, | |
| "eval_runtime": 146.4442, | |
| "eval_samples_per_second": 38.622, | |
| "eval_steps_per_second": 4.828, | |
| "eval_wer": 0.7589510680297219, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1548586914440573, | |
| "eval_loss": 1.0144398212432861, | |
| "eval_runtime": 146.6444, | |
| "eval_samples_per_second": 38.569, | |
| "eval_steps_per_second": 4.821, | |
| "eval_wer": 0.7211567780969652, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17421602787456447, | |
| "eval_loss": 0.9338767528533936, | |
| "eval_runtime": 147.2439, | |
| "eval_samples_per_second": 38.412, | |
| "eval_steps_per_second": 4.802, | |
| "eval_wer": 0.6931520919259841, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "grad_norm": 3.122023105621338, | |
| "learning_rate": 0.0002843684210526315, | |
| "loss": 1.0454, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "eval_loss": 0.8805950284004211, | |
| "eval_runtime": 146.7917, | |
| "eval_samples_per_second": 38.531, | |
| "eval_steps_per_second": 4.816, | |
| "eval_wer": 0.659739050889891, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2129307007355788, | |
| "eval_loss": 0.864371657371521, | |
| "eval_runtime": 146.75, | |
| "eval_samples_per_second": 38.542, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.6553738505239846, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23228803716608595, | |
| "eval_loss": 0.845355749130249, | |
| "eval_runtime": 147.3826, | |
| "eval_samples_per_second": 38.376, | |
| "eval_steps_per_second": 4.797, | |
| "eval_wer": 0.6313652485114988, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.2516453735965931, | |
| "eval_loss": 0.8092750906944275, | |
| "eval_runtime": 147.8588, | |
| "eval_samples_per_second": 38.253, | |
| "eval_steps_per_second": 4.782, | |
| "eval_wer": 0.5919179599107701, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27100271002710025, | |
| "eval_loss": 0.8075847029685974, | |
| "eval_runtime": 147.3238, | |
| "eval_samples_per_second": 38.392, | |
| "eval_steps_per_second": 4.799, | |
| "eval_wer": 0.6072282582529569, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "grad_norm": 3.26877498626709, | |
| "learning_rate": 0.000268578947368421, | |
| "loss": 0.842, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "eval_loss": 0.7783301472663879, | |
| "eval_runtime": 151.3584, | |
| "eval_samples_per_second": 37.368, | |
| "eval_steps_per_second": 4.671, | |
| "eval_wer": 0.5857232270385646, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3097173828881146, | |
| "eval_loss": 0.7964575290679932, | |
| "eval_runtime": 147.3418, | |
| "eval_samples_per_second": 38.387, | |
| "eval_steps_per_second": 4.798, | |
| "eval_wer": 0.5940524145014524, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.32907471931862176, | |
| "eval_loss": 0.7414730191230774, | |
| "eval_runtime": 145.4997, | |
| "eval_samples_per_second": 38.873, | |
| "eval_steps_per_second": 4.859, | |
| "eval_wer": 0.5505127505576864, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.34843205574912894, | |
| "eval_loss": 0.7440277934074402, | |
| "eval_runtime": 144.5466, | |
| "eval_samples_per_second": 39.129, | |
| "eval_steps_per_second": 4.891, | |
| "eval_wer": 0.563656497247677, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.3677893921796361, | |
| "eval_loss": 0.7360929846763611, | |
| "eval_runtime": 144.6106, | |
| "eval_samples_per_second": 39.112, | |
| "eval_steps_per_second": 4.889, | |
| "eval_wer": 0.5865256535764151, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "grad_norm": 1.8907363414764404, | |
| "learning_rate": 0.0002527894736842105, | |
| "loss": 0.755, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "eval_loss": 0.7313781380653381, | |
| "eval_runtime": 144.7461, | |
| "eval_samples_per_second": 39.075, | |
| "eval_steps_per_second": 4.884, | |
| "eval_wer": 0.542729213140537, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4065040650406504, | |
| "eval_loss": 0.6866404414176941, | |
| "eval_runtime": 144.8178, | |
| "eval_samples_per_second": 39.056, | |
| "eval_steps_per_second": 4.882, | |
| "eval_wer": 0.5181428640207989, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4258614014711576, | |
| "eval_loss": 0.6947888731956482, | |
| "eval_runtime": 144.3916, | |
| "eval_samples_per_second": 39.171, | |
| "eval_steps_per_second": 4.896, | |
| "eval_wer": 0.5425847763637239, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.4452187379016647, | |
| "eval_loss": 0.6796152591705322, | |
| "eval_runtime": 144.4575, | |
| "eval_samples_per_second": 39.153, | |
| "eval_steps_per_second": 4.894, | |
| "eval_wer": 0.5158960697148176, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.4645760743321719, | |
| "eval_loss": 0.6898629069328308, | |
| "eval_runtime": 144.5493, | |
| "eval_samples_per_second": 39.129, | |
| "eval_steps_per_second": 4.891, | |
| "eval_wer": 0.5305162812344529, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "grad_norm": 2.9580163955688477, | |
| "learning_rate": 0.000237, | |
| "loss": 0.6884, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "eval_loss": 0.6736070513725281, | |
| "eval_runtime": 144.9197, | |
| "eval_samples_per_second": 39.029, | |
| "eval_steps_per_second": 4.879, | |
| "eval_wer": 0.5103272295421354, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5032907471931862, | |
| "eval_loss": 0.6728157997131348, | |
| "eval_runtime": 145.4712, | |
| "eval_samples_per_second": 38.881, | |
| "eval_steps_per_second": 4.86, | |
| "eval_wer": 0.5257338190688643, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5226480836236934, | |
| "eval_loss": 0.6537250876426697, | |
| "eval_runtime": 145.152, | |
| "eval_samples_per_second": 38.966, | |
| "eval_steps_per_second": 4.871, | |
| "eval_wer": 0.5026560318402851, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5420054200542005, | |
| "eval_loss": 0.631415843963623, | |
| "eval_runtime": 145.09, | |
| "eval_samples_per_second": 38.983, | |
| "eval_steps_per_second": 4.873, | |
| "eval_wer": 0.48227439777888337, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.5613627564847077, | |
| "eval_loss": 0.6316519379615784, | |
| "eval_runtime": 144.7345, | |
| "eval_samples_per_second": 39.078, | |
| "eval_steps_per_second": 4.885, | |
| "eval_wer": 0.48296448460143476, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "grad_norm": 8.703415870666504, | |
| "learning_rate": 0.00022121052631578946, | |
| "loss": 0.6756, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "eval_loss": 0.6204401254653931, | |
| "eval_runtime": 144.9451, | |
| "eval_samples_per_second": 39.022, | |
| "eval_steps_per_second": 4.878, | |
| "eval_wer": 0.4761117619681918, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6000774293457221, | |
| "eval_loss": 0.6310548782348633, | |
| "eval_runtime": 144.6818, | |
| "eval_samples_per_second": 39.093, | |
| "eval_steps_per_second": 4.887, | |
| "eval_wer": 0.4810868065028647, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.6194347657762292, | |
| "eval_loss": 0.623622477054596, | |
| "eval_runtime": 145.2962, | |
| "eval_samples_per_second": 38.927, | |
| "eval_steps_per_second": 4.866, | |
| "eval_wer": 0.48631862752964966, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.6387921022067363, | |
| "eval_loss": 0.6224333047866821, | |
| "eval_runtime": 144.731, | |
| "eval_samples_per_second": 39.079, | |
| "eval_steps_per_second": 4.885, | |
| "eval_wer": 0.46287172409365923, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.6581494386372435, | |
| "eval_loss": 0.597332239151001, | |
| "eval_runtime": 145.3577, | |
| "eval_samples_per_second": 38.911, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.4622939769864069, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "grad_norm": 4.671802997589111, | |
| "learning_rate": 0.00020542105263157893, | |
| "loss": 0.6435, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "eval_loss": 0.5913041830062866, | |
| "eval_runtime": 145.2068, | |
| "eval_samples_per_second": 38.951, | |
| "eval_steps_per_second": 4.869, | |
| "eval_wer": 0.47079969828762175, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6968641114982579, | |
| "eval_loss": 0.6086587905883789, | |
| "eval_runtime": 145.0548, | |
| "eval_samples_per_second": 38.992, | |
| "eval_steps_per_second": 4.874, | |
| "eval_wer": 0.4743945691771918, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.716221447928765, | |
| "eval_loss": 0.5827310681343079, | |
| "eval_runtime": 145.6494, | |
| "eval_samples_per_second": 38.833, | |
| "eval_steps_per_second": 4.854, | |
| "eval_wer": 0.45213525701722007, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.7355787843592722, | |
| "eval_loss": 0.5875205397605896, | |
| "eval_runtime": 145.7866, | |
| "eval_samples_per_second": 38.796, | |
| "eval_steps_per_second": 4.85, | |
| "eval_wer": 0.460785415095248, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.7549361207897793, | |
| "eval_loss": 0.592467725276947, | |
| "eval_runtime": 145.4885, | |
| "eval_samples_per_second": 38.876, | |
| "eval_steps_per_second": 4.859, | |
| "eval_wer": 0.45573012790679013, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.7742934572202865, | |
| "grad_norm": 3.854473114013672, | |
| "learning_rate": 0.0001896315789473684, | |
| "loss": 0.6282, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7742934572202865, | |
| "eval_loss": 0.5799296498298645, | |
| "eval_runtime": 145.6866, | |
| "eval_samples_per_second": 38.823, | |
| "eval_steps_per_second": 4.853, | |
| "eval_wer": 0.4494230553192855, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7936507936507936, | |
| "eval_loss": 0.567882239818573, | |
| "eval_runtime": 146.7561, | |
| "eval_samples_per_second": 38.54, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.45260066440917335, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "eval_loss": 0.5699547529220581, | |
| "eval_runtime": 145.6353, | |
| "eval_samples_per_second": 38.837, | |
| "eval_steps_per_second": 4.855, | |
| "eval_wer": 0.4549597984304537, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.832365466511808, | |
| "eval_loss": 0.5610417127609253, | |
| "eval_runtime": 145.0921, | |
| "eval_samples_per_second": 38.982, | |
| "eval_steps_per_second": 4.873, | |
| "eval_wer": 0.43432138787693986, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.8517228029423152, | |
| "eval_loss": 0.5616321563720703, | |
| "eval_runtime": 145.7565, | |
| "eval_samples_per_second": 38.804, | |
| "eval_steps_per_second": 4.851, | |
| "eval_wer": 0.42727608287461283, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.8710801393728222, | |
| "grad_norm": 3.4542479515075684, | |
| "learning_rate": 0.0001738421052631579, | |
| "loss": 0.5937, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8710801393728222, | |
| "eval_loss": 0.5463823080062866, | |
| "eval_runtime": 145.3415, | |
| "eval_samples_per_second": 38.915, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.42207635890934186, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8904374758033294, | |
| "eval_loss": 0.5485692620277405, | |
| "eval_runtime": 145.5247, | |
| "eval_samples_per_second": 38.866, | |
| "eval_steps_per_second": 4.858, | |
| "eval_wer": 0.4287685962350147, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.9097948122338366, | |
| "eval_loss": 0.5307685136795044, | |
| "eval_runtime": 145.423, | |
| "eval_samples_per_second": 38.893, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 0.41673219816725776, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.9291521486643438, | |
| "eval_loss": 0.5520421862602234, | |
| "eval_runtime": 145.5542, | |
| "eval_samples_per_second": 38.858, | |
| "eval_steps_per_second": 4.857, | |
| "eval_wer": 0.41997400138017366, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.948509485094851, | |
| "eval_loss": 0.5321463346481323, | |
| "eval_runtime": 146.3191, | |
| "eval_samples_per_second": 38.655, | |
| "eval_steps_per_second": 4.832, | |
| "eval_wer": 0.41804817768933256, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.9678668215253581, | |
| "grad_norm": 5.257072925567627, | |
| "learning_rate": 0.00015808421052631577, | |
| "loss": 0.5659, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9678668215253581, | |
| "eval_loss": 0.5333205461502075, | |
| "eval_runtime": 145.4188, | |
| "eval_samples_per_second": 38.895, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 0.4176148673588933, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9872241579558653, | |
| "eval_loss": 0.5260410904884338, | |
| "eval_runtime": 146.9543, | |
| "eval_samples_per_second": 38.488, | |
| "eval_steps_per_second": 4.811, | |
| "eval_wer": 0.41113126093306157, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.0065814943863725, | |
| "eval_loss": 0.5185408592224121, | |
| "eval_runtime": 145.8818, | |
| "eval_samples_per_second": 38.771, | |
| "eval_steps_per_second": 4.846, | |
| "eval_wer": 0.3973616215435477, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.0259388308168795, | |
| "eval_loss": 0.5147408843040466, | |
| "eval_runtime": 150.3669, | |
| "eval_samples_per_second": 37.615, | |
| "eval_steps_per_second": 4.702, | |
| "eval_wer": 0.3917927813708655, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.0452961672473868, | |
| "eval_loss": 0.5154542326927185, | |
| "eval_runtime": 145.571, | |
| "eval_samples_per_second": 38.854, | |
| "eval_steps_per_second": 4.857, | |
| "eval_wer": 0.3975863009741458, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.064653503677894, | |
| "grad_norm": 1.1321543455123901, | |
| "learning_rate": 0.00014232631578947366, | |
| "loss": 0.4928, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.064653503677894, | |
| "eval_loss": 0.5057936906814575, | |
| "eval_runtime": 144.9813, | |
| "eval_samples_per_second": 39.012, | |
| "eval_steps_per_second": 4.876, | |
| "eval_wer": 0.3936062653464075, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.084010840108401, | |
| "eval_loss": 0.504751443862915, | |
| "eval_runtime": 145.1866, | |
| "eval_samples_per_second": 38.957, | |
| "eval_steps_per_second": 4.87, | |
| "eval_wer": 0.3964629038211552, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.1033681765389083, | |
| "eval_loss": 0.5011361241340637, | |
| "eval_runtime": 146.2522, | |
| "eval_samples_per_second": 38.673, | |
| "eval_steps_per_second": 4.834, | |
| "eval_wer": 0.3818266437707628, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.1227255129694154, | |
| "eval_loss": 0.4964805543422699, | |
| "eval_runtime": 146.0597, | |
| "eval_samples_per_second": 38.724, | |
| "eval_steps_per_second": 4.84, | |
| "eval_wer": 0.3830463321082955, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.1420828493999227, | |
| "eval_loss": 0.4969277083873749, | |
| "eval_runtime": 145.1404, | |
| "eval_samples_per_second": 38.969, | |
| "eval_steps_per_second": 4.871, | |
| "eval_wer": 0.383993195422959, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.1614401858304297, | |
| "grad_norm": 1.1488393545150757, | |
| "learning_rate": 0.00012653684210526316, | |
| "loss": 0.4619, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1614401858304297, | |
| "eval_loss": 0.4863020181655884, | |
| "eval_runtime": 146.181, | |
| "eval_samples_per_second": 38.692, | |
| "eval_steps_per_second": 4.836, | |
| "eval_wer": 0.3799650142029497, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1807975222609368, | |
| "eval_loss": 0.49075642228126526, | |
| "eval_runtime": 145.2768, | |
| "eval_samples_per_second": 38.933, | |
| "eval_steps_per_second": 4.867, | |
| "eval_wer": 0.37998106273370674, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.2001548586914441, | |
| "eval_loss": 0.4835449457168579, | |
| "eval_runtime": 145.0604, | |
| "eval_samples_per_second": 38.991, | |
| "eval_steps_per_second": 4.874, | |
| "eval_wer": 0.3712185649403797, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "eval_loss": 0.4926937520503998, | |
| "eval_runtime": 145.5282, | |
| "eval_samples_per_second": 38.865, | |
| "eval_steps_per_second": 4.858, | |
| "eval_wer": 0.3766590168670058, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.2388695315524583, | |
| "eval_loss": 0.4942048490047455, | |
| "eval_runtime": 146.7525, | |
| "eval_samples_per_second": 38.541, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.368281683811847, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.2582268679829656, | |
| "grad_norm": 1.8319953680038452, | |
| "learning_rate": 0.00011074736842105263, | |
| "loss": 0.4421, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2582268679829656, | |
| "eval_loss": 0.4833586513996124, | |
| "eval_runtime": 147.4272, | |
| "eval_samples_per_second": 38.365, | |
| "eval_steps_per_second": 4.796, | |
| "eval_wer": 0.37393076663831426, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2775842044134726, | |
| "eval_loss": 0.47513511776924133, | |
| "eval_runtime": 146.0617, | |
| "eval_samples_per_second": 38.723, | |
| "eval_steps_per_second": 4.84, | |
| "eval_wer": 0.3633868819309592, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.29694154084398, | |
| "eval_loss": 0.4733775854110718, | |
| "eval_runtime": 145.526, | |
| "eval_samples_per_second": 38.866, | |
| "eval_steps_per_second": 4.858, | |
| "eval_wer": 0.36327454221566013, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.316298877274487, | |
| "eval_loss": 0.4685443937778473, | |
| "eval_runtime": 145.3437, | |
| "eval_samples_per_second": 38.915, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.36447818202243587, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.3356562137049943, | |
| "eval_loss": 0.4654460847377777, | |
| "eval_runtime": 145.3646, | |
| "eval_samples_per_second": 38.909, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.3624560671470527, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "grad_norm": 0.9300881624221802, | |
| "learning_rate": 9.49578947368421e-05, | |
| "loss": 0.4304, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "eval_loss": 0.47420966625213623, | |
| "eval_runtime": 145.5058, | |
| "eval_samples_per_second": 38.871, | |
| "eval_steps_per_second": 4.859, | |
| "eval_wer": 0.3615413008939032, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3743708865660085, | |
| "eval_loss": 0.46446511149406433, | |
| "eval_runtime": 145.4472, | |
| "eval_samples_per_second": 38.887, | |
| "eval_steps_per_second": 4.861, | |
| "eval_wer": 0.359567331610791, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.3937282229965158, | |
| "eval_loss": 0.45991310477256775, | |
| "eval_runtime": 145.7318, | |
| "eval_samples_per_second": 38.811, | |
| "eval_steps_per_second": 4.851, | |
| "eval_wer": 0.3593587007109499, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 1.4130855594270229, | |
| "eval_loss": 0.4554171562194824, | |
| "eval_runtime": 145.4195, | |
| "eval_samples_per_second": 38.894, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 0.3555391503907817, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 1.43244289585753, | |
| "eval_loss": 0.457757830619812, | |
| "eval_runtime": 145.3536, | |
| "eval_samples_per_second": 38.912, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.35780199322752004, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 1.4518002322880372, | |
| "grad_norm": 1.012635588645935, | |
| "learning_rate": 7.916842105263156e-05, | |
| "loss": 0.4275, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4518002322880372, | |
| "eval_loss": 0.45184171199798584, | |
| "eval_runtime": 149.8003, | |
| "eval_samples_per_second": 37.757, | |
| "eval_steps_per_second": 4.72, | |
| "eval_wer": 0.3521850074625668, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4711575687185443, | |
| "eval_loss": 0.44799017906188965, | |
| "eval_runtime": 145.3878, | |
| "eval_samples_per_second": 38.903, | |
| "eval_steps_per_second": 4.863, | |
| "eval_wer": 0.35107765884033315, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 1.4905149051490514, | |
| "eval_loss": 0.4465474486351013, | |
| "eval_runtime": 145.7643, | |
| "eval_samples_per_second": 38.802, | |
| "eval_steps_per_second": 4.85, | |
| "eval_wer": 0.3500986984641556, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 1.5098722415795587, | |
| "eval_loss": 0.44539782404899597, | |
| "eval_runtime": 145.5321, | |
| "eval_samples_per_second": 38.864, | |
| "eval_steps_per_second": 4.858, | |
| "eval_wer": 0.34281266550047346, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 1.5292295780100658, | |
| "eval_loss": 0.44271060824394226, | |
| "eval_runtime": 145.0706, | |
| "eval_samples_per_second": 38.988, | |
| "eval_steps_per_second": 4.873, | |
| "eval_wer": 0.34387186853043605, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 1.5485869144405728, | |
| "grad_norm": 1.030893087387085, | |
| "learning_rate": 6.341052631578946e-05, | |
| "loss": 0.4089, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5485869144405728, | |
| "eval_loss": 0.437569797039032, | |
| "eval_runtime": 144.8474, | |
| "eval_samples_per_second": 39.048, | |
| "eval_steps_per_second": 4.881, | |
| "eval_wer": 0.3406942594405482, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5679442508710801, | |
| "eval_loss": 0.4396042823791504, | |
| "eval_runtime": 145.4051, | |
| "eval_samples_per_second": 38.898, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 0.3415127345091557, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 1.5873015873015874, | |
| "eval_loss": 0.4342670738697052, | |
| "eval_runtime": 144.5678, | |
| "eval_samples_per_second": 39.124, | |
| "eval_steps_per_second": 4.89, | |
| "eval_wer": 0.34217072427019307, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 1.6066589237320945, | |
| "eval_loss": 0.4358905851840973, | |
| "eval_runtime": 144.9749, | |
| "eval_samples_per_second": 39.014, | |
| "eval_steps_per_second": 4.877, | |
| "eval_wer": 0.3406461138482772, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "eval_loss": 0.43577995896339417, | |
| "eval_runtime": 144.8242, | |
| "eval_samples_per_second": 39.054, | |
| "eval_steps_per_second": 4.882, | |
| "eval_wer": 0.3373401165123333, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 1.645373596593109, | |
| "grad_norm": 0.9704302549362183, | |
| "learning_rate": 4.762105263157894e-05, | |
| "loss": 0.4005, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.645373596593109, | |
| "eval_loss": 0.4330734312534332, | |
| "eval_runtime": 145.4279, | |
| "eval_samples_per_second": 38.892, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 0.3364895443822118, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.664730933023616, | |
| "eval_loss": 0.4302414059638977, | |
| "eval_runtime": 144.5881, | |
| "eval_samples_per_second": 39.118, | |
| "eval_steps_per_second": 4.89, | |
| "eval_wer": 0.3352859045754361, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 1.684088269454123, | |
| "eval_loss": 0.43084925413131714, | |
| "eval_runtime": 145.3335, | |
| "eval_samples_per_second": 38.917, | |
| "eval_steps_per_second": 4.865, | |
| "eval_wer": 0.33549453547527724, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 1.7034456058846303, | |
| "eval_loss": 0.4258446991443634, | |
| "eval_runtime": 145.2324, | |
| "eval_samples_per_second": 38.944, | |
| "eval_steps_per_second": 4.868, | |
| "eval_wer": 0.335109370737109, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 1.7228029423151374, | |
| "eval_loss": 0.42223912477493286, | |
| "eval_runtime": 145.1268, | |
| "eval_samples_per_second": 38.973, | |
| "eval_steps_per_second": 4.872, | |
| "eval_wer": 0.33530195310619315, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 1.7421602787456445, | |
| "grad_norm": 1.2587796449661255, | |
| "learning_rate": 3.186315789473684e-05, | |
| "loss": 0.3879, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7421602787456445, | |
| "eval_loss": 0.4238055944442749, | |
| "eval_runtime": 145.2053, | |
| "eval_samples_per_second": 38.952, | |
| "eval_steps_per_second": 4.869, | |
| "eval_wer": 0.33119352923239875, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7615176151761518, | |
| "eval_loss": 0.4244863986968994, | |
| "eval_runtime": 145.3408, | |
| "eval_samples_per_second": 38.915, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.3288022981496044, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 1.7808749516066589, | |
| "eval_loss": 0.42055588960647583, | |
| "eval_runtime": 145.738, | |
| "eval_samples_per_second": 38.809, | |
| "eval_steps_per_second": 4.851, | |
| "eval_wer": 0.32641106706681006, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.800232288037166, | |
| "eval_loss": 0.4201093018054962, | |
| "eval_runtime": 145.1603, | |
| "eval_samples_per_second": 38.964, | |
| "eval_steps_per_second": 4.87, | |
| "eval_wer": 0.3284492304729502, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 1.8195896244676733, | |
| "eval_loss": 0.41892749071121216, | |
| "eval_runtime": 150.4269, | |
| "eval_samples_per_second": 37.6, | |
| "eval_steps_per_second": 4.7, | |
| "eval_wer": 0.324597583091268, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.8389469608981805, | |
| "grad_norm": 1.2362818717956543, | |
| "learning_rate": 1.6073684210526313e-05, | |
| "loss": 0.369, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8389469608981805, | |
| "eval_loss": 0.4160093665122986, | |
| "eval_runtime": 145.5064, | |
| "eval_samples_per_second": 38.871, | |
| "eval_steps_per_second": 4.859, | |
| "eval_wer": 0.32576912583652967, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8583042973286876, | |
| "eval_loss": 0.4141674041748047, | |
| "eval_runtime": 145.8993, | |
| "eval_samples_per_second": 38.766, | |
| "eval_steps_per_second": 4.846, | |
| "eval_wer": 0.3248062139911091, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.8776616337591947, | |
| "eval_loss": 0.41305750608444214, | |
| "eval_runtime": 146.4678, | |
| "eval_samples_per_second": 38.616, | |
| "eval_steps_per_second": 4.827, | |
| "eval_wer": 0.32517533019852035, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 1.897018970189702, | |
| "eval_loss": 0.4127916693687439, | |
| "eval_runtime": 145.2897, | |
| "eval_samples_per_second": 38.929, | |
| "eval_steps_per_second": 4.866, | |
| "eval_wer": 0.32276805058496894, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.916376306620209, | |
| "eval_loss": 0.41222211718559265, | |
| "eval_runtime": 145.7491, | |
| "eval_samples_per_second": 38.806, | |
| "eval_steps_per_second": 4.851, | |
| "eval_wer": 0.3221421578854456, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "grad_norm": 1.1011325120925903, | |
| "learning_rate": 2.842105263157894e-07, | |
| "loss": 0.3738, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "eval_loss": 0.4121682941913605, | |
| "eval_runtime": 145.6809, | |
| "eval_samples_per_second": 38.825, | |
| "eval_steps_per_second": 4.853, | |
| "eval_wer": 0.3222865946622587, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "step": 10000, | |
| "total_flos": 1.1255918428180738e+19, | |
| "train_loss": 0.7340839981079101, | |
| "train_runtime": 18566.2158, | |
| "train_samples_per_second": 4.309, | |
| "train_steps_per_second": 0.539 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1255918428180738e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |