| { | |
| "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": 3.547485828399658, | |
| "eval_runtime": 144.8623, | |
| "eval_samples_per_second": 39.044, | |
| "eval_steps_per_second": 4.88, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03871467286101433, | |
| "eval_loss": 3.0259251594543457, | |
| "eval_runtime": 142.9174, | |
| "eval_samples_per_second": 39.575, | |
| "eval_steps_per_second": 4.947, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05807200929152149, | |
| "eval_loss": 3.0886833667755127, | |
| "eval_runtime": 141.7177, | |
| "eval_samples_per_second": 39.91, | |
| "eval_steps_per_second": 4.989, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07742934572202866, | |
| "eval_loss": 2.3821566104888916, | |
| "eval_runtime": 143.4279, | |
| "eval_samples_per_second": 39.434, | |
| "eval_steps_per_second": 4.929, | |
| "eval_wer": 0.9971915071175234, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "grad_norm": 3.033390760421753, | |
| "learning_rate": 0.0002982, | |
| "loss": 4.0938, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "eval_loss": 1.4546788930892944, | |
| "eval_runtime": 142.7727, | |
| "eval_samples_per_second": 39.615, | |
| "eval_steps_per_second": 4.952, | |
| "eval_wer": 0.9020076711977019, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11614401858304298, | |
| "eval_loss": 1.2602813243865967, | |
| "eval_runtime": 143.0991, | |
| "eval_samples_per_second": 39.525, | |
| "eval_steps_per_second": 4.941, | |
| "eval_wer": 0.8509733433904126, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13550135501355012, | |
| "eval_loss": 1.0939536094665527, | |
| "eval_runtime": 145.4158, | |
| "eval_samples_per_second": 38.895, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 0.7654667715170677, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1548586914440573, | |
| "eval_loss": 1.0704576969146729, | |
| "eval_runtime": 148.1277, | |
| "eval_samples_per_second": 38.183, | |
| "eval_steps_per_second": 4.773, | |
| "eval_wer": 0.7601547078364975, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17421602787456447, | |
| "eval_loss": 0.9356458187103271, | |
| "eval_runtime": 143.6296, | |
| "eval_samples_per_second": 39.379, | |
| "eval_steps_per_second": 4.922, | |
| "eval_wer": 0.6972926128612925, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "grad_norm": 3.2989861965179443, | |
| "learning_rate": 0.0002843684210526315, | |
| "loss": 1.0597, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "eval_loss": 0.9103516936302185, | |
| "eval_runtime": 146.7237, | |
| "eval_samples_per_second": 38.549, | |
| "eval_steps_per_second": 4.819, | |
| "eval_wer": 0.6765579111232367, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2129307007355788, | |
| "eval_loss": 0.8879104256629944, | |
| "eval_runtime": 153.6385, | |
| "eval_samples_per_second": 36.814, | |
| "eval_steps_per_second": 4.602, | |
| "eval_wer": 0.6569947521304424, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23228803716608595, | |
| "eval_loss": 0.8594633936882019, | |
| "eval_runtime": 147.9212, | |
| "eval_samples_per_second": 38.237, | |
| "eval_steps_per_second": 4.78, | |
| "eval_wer": 0.6611834186580219, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.2516453735965931, | |
| "eval_loss": 0.8351845145225525, | |
| "eval_runtime": 148.1861, | |
| "eval_samples_per_second": 38.168, | |
| "eval_steps_per_second": 4.771, | |
| "eval_wer": 0.6075331803373402, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27100271002710025, | |
| "eval_loss": 0.791232168674469, | |
| "eval_runtime": 148.9749, | |
| "eval_samples_per_second": 37.966, | |
| "eval_steps_per_second": 4.746, | |
| "eval_wer": 0.6033124167482466, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "grad_norm": 6.474522113800049, | |
| "learning_rate": 0.000268578947368421, | |
| "loss": 0.8484, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "eval_loss": 0.7862286567687988, | |
| "eval_runtime": 146.7521, | |
| "eval_samples_per_second": 38.541, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.6067468023302467, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3097173828881146, | |
| "eval_loss": 0.7790109515190125, | |
| "eval_runtime": 147.7062, | |
| "eval_samples_per_second": 38.292, | |
| "eval_steps_per_second": 4.787, | |
| "eval_wer": 0.6009051371346953, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.32907471931862176, | |
| "eval_loss": 0.7678210735321045, | |
| "eval_runtime": 148.3951, | |
| "eval_samples_per_second": 38.114, | |
| "eval_steps_per_second": 4.764, | |
| "eval_wer": 0.5629182648328546, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.34843205574912894, | |
| "eval_loss": 0.7514644861221313, | |
| "eval_runtime": 149.2674, | |
| "eval_samples_per_second": 37.892, | |
| "eval_steps_per_second": 4.736, | |
| "eval_wer": 0.5798655133122562, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.3677893921796361, | |
| "eval_loss": 0.7423551678657532, | |
| "eval_runtime": 149.3427, | |
| "eval_samples_per_second": 37.873, | |
| "eval_steps_per_second": 4.734, | |
| "eval_wer": 0.5859158094076488, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "grad_norm": 2.573913335800171, | |
| "learning_rate": 0.0002527894736842105, | |
| "loss": 0.764, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "eval_loss": 0.7129915356636047, | |
| "eval_runtime": 148.3711, | |
| "eval_samples_per_second": 38.121, | |
| "eval_steps_per_second": 4.765, | |
| "eval_wer": 0.5520855065718734, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4065040650406504, | |
| "eval_loss": 0.7114368677139282, | |
| "eval_runtime": 148.2007, | |
| "eval_samples_per_second": 38.164, | |
| "eval_steps_per_second": 4.771, | |
| "eval_wer": 0.5407712923881819, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4258614014711576, | |
| "eval_loss": 0.7228682637214661, | |
| "eval_runtime": 149.1432, | |
| "eval_samples_per_second": 37.923, | |
| "eval_steps_per_second": 4.74, | |
| "eval_wer": 0.5577024923368266, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.4452187379016647, | |
| "eval_loss": 0.6773180961608887, | |
| "eval_runtime": 154.221, | |
| "eval_samples_per_second": 36.675, | |
| "eval_steps_per_second": 4.584, | |
| "eval_wer": 0.5160084094301167, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.4645760743321719, | |
| "eval_loss": 0.6784498691558838, | |
| "eval_runtime": 149.2744, | |
| "eval_samples_per_second": 37.89, | |
| "eval_steps_per_second": 4.736, | |
| "eval_wer": 0.5177897963441447, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "grad_norm": 3.0869553089141846, | |
| "learning_rate": 0.000237, | |
| "loss": 0.6868, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "eval_loss": 0.672030508518219, | |
| "eval_runtime": 149.4453, | |
| "eval_samples_per_second": 37.847, | |
| "eval_steps_per_second": 4.731, | |
| "eval_wer": 0.5261831779300605, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5032907471931862, | |
| "eval_loss": 0.6804332137107849, | |
| "eval_runtime": 151.0327, | |
| "eval_samples_per_second": 37.449, | |
| "eval_steps_per_second": 4.681, | |
| "eval_wer": 0.5336617932628268, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5226480836236934, | |
| "eval_loss": 0.6598911285400391, | |
| "eval_runtime": 149.0299, | |
| "eval_samples_per_second": 37.952, | |
| "eval_steps_per_second": 4.744, | |
| "eval_wer": 0.5023832068174159, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5420054200542005, | |
| "eval_loss": 0.6287100911140442, | |
| "eval_runtime": 149.9845, | |
| "eval_samples_per_second": 37.711, | |
| "eval_steps_per_second": 4.714, | |
| "eval_wer": 0.4902023719728459, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.5613627564847077, | |
| "eval_loss": 0.6304338574409485, | |
| "eval_runtime": 150.016, | |
| "eval_samples_per_second": 37.703, | |
| "eval_steps_per_second": 4.713, | |
| "eval_wer": 0.49471200911556545, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "grad_norm": 5.678714275360107, | |
| "learning_rate": 0.00022121052631578946, | |
| "loss": 0.6761, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "eval_loss": 0.6258472204208374, | |
| "eval_runtime": 149.8088, | |
| "eval_samples_per_second": 37.755, | |
| "eval_steps_per_second": 4.719, | |
| "eval_wer": 0.48513103625363097, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6000774293457221, | |
| "eval_loss": 0.6310975551605225, | |
| "eval_runtime": 148.9286, | |
| "eval_samples_per_second": 37.978, | |
| "eval_steps_per_second": 4.747, | |
| "eval_wer": 0.4989809182969299, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.6194347657762292, | |
| "eval_loss": 0.6171565651893616, | |
| "eval_runtime": 148.6924, | |
| "eval_samples_per_second": 38.038, | |
| "eval_steps_per_second": 4.755, | |
| "eval_wer": 0.4901060807883038, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.6387921022067363, | |
| "eval_loss": 0.6187321543693542, | |
| "eval_runtime": 149.7679, | |
| "eval_samples_per_second": 37.765, | |
| "eval_steps_per_second": 4.721, | |
| "eval_wer": 0.46661103176004237, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.6581494386372435, | |
| "eval_loss": 0.6044796109199524, | |
| "eval_runtime": 149.5983, | |
| "eval_samples_per_second": 37.808, | |
| "eval_steps_per_second": 4.726, | |
| "eval_wer": 0.4725489881401358, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "grad_norm": 4.122500419616699, | |
| "learning_rate": 0.00020542105263157893, | |
| "loss": 0.6462, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "eval_loss": 0.5950499773025513, | |
| "eval_runtime": 148.2511, | |
| "eval_samples_per_second": 38.151, | |
| "eval_steps_per_second": 4.769, | |
| "eval_wer": 0.4716823674792573, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6968641114982579, | |
| "eval_loss": 0.5902624726295471, | |
| "eval_runtime": 149.3094, | |
| "eval_samples_per_second": 37.881, | |
| "eval_steps_per_second": 4.735, | |
| "eval_wer": 0.4602237165187527, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.716221447928765, | |
| "eval_loss": 0.5864866375923157, | |
| "eval_runtime": 149.6434, | |
| "eval_samples_per_second": 37.797, | |
| "eval_steps_per_second": 4.725, | |
| "eval_wer": 0.47267737638619184, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.7355787843592722, | |
| "eval_loss": 0.5820363759994507, | |
| "eval_runtime": 148.886, | |
| "eval_samples_per_second": 37.989, | |
| "eval_steps_per_second": 4.749, | |
| "eval_wer": 0.459036125242734, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.7549361207897793, | |
| "eval_loss": 0.6025602221488953, | |
| "eval_runtime": 148.9627, | |
| "eval_samples_per_second": 37.969, | |
| "eval_steps_per_second": 4.746, | |
| "eval_wer": 0.48296448460143476, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.7742934572202865, | |
| "grad_norm": 5.146019458770752, | |
| "learning_rate": 0.0001896315789473684, | |
| "loss": 0.6193, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7742934572202865, | |
| "eval_loss": 0.5807139277458191, | |
| "eval_runtime": 147.9966, | |
| "eval_samples_per_second": 38.217, | |
| "eval_steps_per_second": 4.777, | |
| "eval_wer": 0.44963168621912664, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7936507936507936, | |
| "eval_loss": 0.5620962977409363, | |
| "eval_runtime": 148.8391, | |
| "eval_samples_per_second": 38.001, | |
| "eval_steps_per_second": 4.75, | |
| "eval_wer": 0.44857248318916404, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "eval_loss": 0.5730157494544983, | |
| "eval_runtime": 148.9808, | |
| "eval_samples_per_second": 37.965, | |
| "eval_steps_per_second": 4.746, | |
| "eval_wer": 0.4593410473271172, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.832365466511808, | |
| "eval_loss": 0.5592055916786194, | |
| "eval_runtime": 147.8897, | |
| "eval_samples_per_second": 38.245, | |
| "eval_steps_per_second": 4.781, | |
| "eval_wer": 0.43741875431304267, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.8517228029423152, | |
| "eval_loss": 0.5621338486671448, | |
| "eval_runtime": 148.7799, | |
| "eval_samples_per_second": 38.016, | |
| "eval_steps_per_second": 4.752, | |
| "eval_wer": 0.42387379435412686, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.8710801393728222, | |
| "grad_norm": 2.8218295574188232, | |
| "learning_rate": 0.0001738421052631579, | |
| "loss": 0.59, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8710801393728222, | |
| "eval_loss": 0.545798659324646, | |
| "eval_runtime": 150.2397, | |
| "eval_samples_per_second": 37.647, | |
| "eval_steps_per_second": 4.706, | |
| "eval_wer": 0.4304055463722296, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8904374758033294, | |
| "eval_loss": 0.5406409502029419, | |
| "eval_runtime": 148.931, | |
| "eval_samples_per_second": 37.977, | |
| "eval_steps_per_second": 4.747, | |
| "eval_wer": 0.4270674519747717, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.9097948122338366, | |
| "eval_loss": 0.5268651247024536, | |
| "eval_runtime": 148.7725, | |
| "eval_samples_per_second": 38.018, | |
| "eval_steps_per_second": 4.752, | |
| "eval_wer": 0.41315337580844474, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.9291521486643438, | |
| "eval_loss": 0.5362106561660767, | |
| "eval_runtime": 147.8165, | |
| "eval_samples_per_second": 38.264, | |
| "eval_steps_per_second": 4.783, | |
| "eval_wer": 0.4214665147405755, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.948509485094851, | |
| "eval_loss": 0.5226009488105774, | |
| "eval_runtime": 149.4387, | |
| "eval_samples_per_second": 37.848, | |
| "eval_steps_per_second": 4.731, | |
| "eval_wer": 0.41626679077530454, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.9678668215253581, | |
| "grad_norm": 7.241621017456055, | |
| "learning_rate": 0.00015808421052631577, | |
| "loss": 0.5636, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9678668215253581, | |
| "eval_loss": 0.5297274589538574, | |
| "eval_runtime": 149.3726, | |
| "eval_samples_per_second": 37.865, | |
| "eval_steps_per_second": 4.733, | |
| "eval_wer": 0.4148384715379307, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9872241579558653, | |
| "eval_loss": 0.5225785970687866, | |
| "eval_runtime": 149.3039, | |
| "eval_samples_per_second": 37.882, | |
| "eval_steps_per_second": 4.735, | |
| "eval_wer": 0.413634831731155, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.0065814943863725, | |
| "eval_loss": 0.5239331722259521, | |
| "eval_runtime": 149.3008, | |
| "eval_samples_per_second": 37.883, | |
| "eval_steps_per_second": 4.735, | |
| "eval_wer": 0.4054179839835663, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.0259388308168795, | |
| "eval_loss": 0.5382751226425171, | |
| "eval_runtime": 148.0837, | |
| "eval_samples_per_second": 38.195, | |
| "eval_steps_per_second": 4.774, | |
| "eval_wer": 0.4057871001909775, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.0452961672473868, | |
| "eval_loss": 0.5125272274017334, | |
| "eval_runtime": 149.1205, | |
| "eval_samples_per_second": 37.929, | |
| "eval_steps_per_second": 4.741, | |
| "eval_wer": 0.4067179149748841, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.064653503677894, | |
| "grad_norm": 1.124423623085022, | |
| "learning_rate": 0.00014232631578947366, | |
| "loss": 0.4924, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.064653503677894, | |
| "eval_loss": 0.5029215812683105, | |
| "eval_runtime": 147.7988, | |
| "eval_samples_per_second": 38.268, | |
| "eval_steps_per_second": 4.784, | |
| "eval_wer": 0.39533950666816453, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.084010840108401, | |
| "eval_loss": 0.505442202091217, | |
| "eval_runtime": 149.3815, | |
| "eval_samples_per_second": 37.863, | |
| "eval_steps_per_second": 4.733, | |
| "eval_wer": 0.3932050520774823, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.1033681765389083, | |
| "eval_loss": 0.4968744218349457, | |
| "eval_runtime": 149.7956, | |
| "eval_samples_per_second": 37.758, | |
| "eval_steps_per_second": 4.72, | |
| "eval_wer": 0.3894015502880711, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.1227255129694154, | |
| "eval_loss": 0.49354633688926697, | |
| "eval_runtime": 148.4196, | |
| "eval_samples_per_second": 38.108, | |
| "eval_steps_per_second": 4.764, | |
| "eval_wer": 0.38508449551443563, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.1420828493999227, | |
| "eval_loss": 0.49766939878463745, | |
| "eval_runtime": 149.5302, | |
| "eval_samples_per_second": 37.825, | |
| "eval_steps_per_second": 4.728, | |
| "eval_wer": 0.3816501099324357, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.1614401858304297, | |
| "grad_norm": 1.9677255153656006, | |
| "learning_rate": 0.00012653684210526316, | |
| "loss": 0.4602, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1614401858304297, | |
| "eval_loss": 0.4862758219242096, | |
| "eval_runtime": 150.9135, | |
| "eval_samples_per_second": 37.478, | |
| "eval_steps_per_second": 4.685, | |
| "eval_wer": 0.387395483943445, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1807975222609368, | |
| "eval_loss": 0.4906172454357147, | |
| "eval_runtime": 148.5353, | |
| "eval_samples_per_second": 38.078, | |
| "eval_steps_per_second": 4.76, | |
| "eval_wer": 0.3776700743046974, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.2001548586914441, | |
| "eval_loss": 0.4891129434108734, | |
| "eval_runtime": 149.6289, | |
| "eval_samples_per_second": 37.8, | |
| "eval_steps_per_second": 4.725, | |
| "eval_wer": 0.3763861918441367, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "eval_loss": 0.488125741481781, | |
| "eval_runtime": 148.288, | |
| "eval_samples_per_second": 38.142, | |
| "eval_steps_per_second": 4.768, | |
| "eval_wer": 0.3800934024490058, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.2388695315524583, | |
| "eval_loss": 0.48135778307914734, | |
| "eval_runtime": 147.6048, | |
| "eval_samples_per_second": 38.319, | |
| "eval_steps_per_second": 4.79, | |
| "eval_wer": 0.37266293270851053, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.2582268679829656, | |
| "grad_norm": 1.2907174825668335, | |
| "learning_rate": 0.00011074736842105263, | |
| "loss": 0.4407, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2582268679829656, | |
| "eval_loss": 0.47142112255096436, | |
| "eval_runtime": 147.9064, | |
| "eval_samples_per_second": 38.24, | |
| "eval_steps_per_second": 4.78, | |
| "eval_wer": 0.37723676397425815, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2775842044134726, | |
| "eval_loss": 0.47389352321624756, | |
| "eval_runtime": 146.8545, | |
| "eval_samples_per_second": 38.514, | |
| "eval_steps_per_second": 4.814, | |
| "eval_wer": 0.3705605751793423, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.29694154084398, | |
| "eval_loss": 0.4691925644874573, | |
| "eval_runtime": 146.6568, | |
| "eval_samples_per_second": 38.566, | |
| "eval_steps_per_second": 4.821, | |
| "eval_wer": 0.3713790502479498, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.316298877274487, | |
| "eval_loss": 0.4672953486442566, | |
| "eval_runtime": 146.8165, | |
| "eval_samples_per_second": 38.524, | |
| "eval_steps_per_second": 4.816, | |
| "eval_wer": 0.3728073694853236, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.3356562137049943, | |
| "eval_loss": 0.46098417043685913, | |
| "eval_runtime": 147.3051, | |
| "eval_samples_per_second": 38.397, | |
| "eval_steps_per_second": 4.8, | |
| "eval_wer": 0.36780022788913674, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "grad_norm": 0.8472552299499512, | |
| "learning_rate": 9.49578947368421e-05, | |
| "loss": 0.4284, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "eval_loss": 0.47299668192863464, | |
| "eval_runtime": 151.323, | |
| "eval_samples_per_second": 37.377, | |
| "eval_steps_per_second": 4.672, | |
| "eval_wer": 0.36531270562180035, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3743708865660085, | |
| "eval_loss": 0.46056076884269714, | |
| "eval_runtime": 146.0139, | |
| "eval_samples_per_second": 38.736, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.36399672609972555, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.3937282229965158, | |
| "eval_loss": 0.4571812152862549, | |
| "eval_runtime": 146.7792, | |
| "eval_samples_per_second": 38.534, | |
| "eval_steps_per_second": 4.817, | |
| "eval_wer": 0.3620067082858564, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 1.4130855594270229, | |
| "eval_loss": 0.45746785402297974, | |
| "eval_runtime": 146.8097, | |
| "eval_samples_per_second": 38.526, | |
| "eval_steps_per_second": 4.816, | |
| "eval_wer": 0.362969620131277, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 1.43244289585753, | |
| "eval_loss": 0.45778077840805054, | |
| "eval_runtime": 146.9433, | |
| "eval_samples_per_second": 38.491, | |
| "eval_steps_per_second": 4.811, | |
| "eval_wer": 0.3590216815650527, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 1.4518002322880372, | |
| "grad_norm": 0.9635696411132812, | |
| "learning_rate": 7.916842105263156e-05, | |
| "loss": 0.4299, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4518002322880372, | |
| "eval_loss": 0.4477390646934509, | |
| "eval_runtime": 146.7454, | |
| "eval_samples_per_second": 38.543, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.3569193240358845, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4711575687185443, | |
| "eval_loss": 0.4441732168197632, | |
| "eval_runtime": 147.4263, | |
| "eval_samples_per_second": 38.365, | |
| "eval_steps_per_second": 4.796, | |
| "eval_wer": 0.3551700341833705, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 1.4905149051490514, | |
| "eval_loss": 0.4420062303543091, | |
| "eval_runtime": 146.725, | |
| "eval_samples_per_second": 38.548, | |
| "eval_steps_per_second": 4.819, | |
| "eval_wer": 0.3546083356068752, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 1.5098722415795587, | |
| "eval_loss": 0.4436999559402466, | |
| "eval_runtime": 145.7818, | |
| "eval_samples_per_second": 38.798, | |
| "eval_steps_per_second": 4.85, | |
| "eval_wer": 0.3482531174270995, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 1.5292295780100658, | |
| "eval_loss": 0.43728071451187134, | |
| "eval_runtime": 146.721, | |
| "eval_samples_per_second": 38.549, | |
| "eval_steps_per_second": 4.819, | |
| "eval_wer": 0.3485740880422397, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 1.5485869144405728, | |
| "grad_norm": 1.1358890533447266, | |
| "learning_rate": 6.341052631578946e-05, | |
| "loss": 0.408, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5485869144405728, | |
| "eval_loss": 0.4335756301879883, | |
| "eval_runtime": 146.7599, | |
| "eval_samples_per_second": 38.539, | |
| "eval_steps_per_second": 4.817, | |
| "eval_wer": 0.3464075363900435, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5679442508710801, | |
| "eval_loss": 0.4347936511039734, | |
| "eval_runtime": 146.7423, | |
| "eval_samples_per_second": 38.544, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.34475453772207154, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 1.5873015873015874, | |
| "eval_loss": 0.42762240767478943, | |
| "eval_runtime": 151.2432, | |
| "eval_samples_per_second": 37.397, | |
| "eval_steps_per_second": 4.675, | |
| "eval_wer": 0.34180160806278187, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 1.6066589237320945, | |
| "eval_loss": 0.42939648032188416, | |
| "eval_runtime": 146.0228, | |
| "eval_samples_per_second": 38.734, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.3399078814334548, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "eval_loss": 0.42716294527053833, | |
| "eval_runtime": 145.901, | |
| "eval_samples_per_second": 38.766, | |
| "eval_steps_per_second": 4.846, | |
| "eval_wer": 0.3387523872189501, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 1.645373596593109, | |
| "grad_norm": 1.037522792816162, | |
| "learning_rate": 4.762105263157894e-05, | |
| "loss": 0.3964, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.645373596593109, | |
| "eval_loss": 0.4310940206050873, | |
| "eval_runtime": 145.7507, | |
| "eval_samples_per_second": 38.806, | |
| "eval_steps_per_second": 4.851, | |
| "eval_wer": 0.3408707932788753, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.664730933023616, | |
| "eval_loss": 0.4260464608669281, | |
| "eval_runtime": 146.3966, | |
| "eval_samples_per_second": 38.635, | |
| "eval_steps_per_second": 4.829, | |
| "eval_wer": 0.3381264945194267, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 1.684088269454123, | |
| "eval_loss": 0.4260489046573639, | |
| "eval_runtime": 146.6331, | |
| "eval_samples_per_second": 38.572, | |
| "eval_steps_per_second": 4.822, | |
| "eval_wer": 0.3370672914894641, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 1.7034456058846303, | |
| "eval_loss": 0.4259546101093292, | |
| "eval_runtime": 146.1762, | |
| "eval_samples_per_second": 38.693, | |
| "eval_steps_per_second": 4.837, | |
| "eval_wer": 0.33636115613615575, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 1.7228029423151374, | |
| "eval_loss": 0.42149877548217773, | |
| "eval_runtime": 147.5316, | |
| "eval_samples_per_second": 38.338, | |
| "eval_steps_per_second": 4.792, | |
| "eval_wer": 0.335109370737109, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 1.7421602787456445, | |
| "grad_norm": 1.4968059062957764, | |
| "learning_rate": 3.186315789473684e-05, | |
| "loss": 0.3866, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7421602787456445, | |
| "eval_loss": 0.4234353303909302, | |
| "eval_runtime": 146.8779, | |
| "eval_samples_per_second": 38.508, | |
| "eval_steps_per_second": 4.814, | |
| "eval_wer": 0.3330391102694548, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7615176151761518, | |
| "eval_loss": 0.4210032522678375, | |
| "eval_runtime": 146.0169, | |
| "eval_samples_per_second": 38.735, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.3318515189934362, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 1.7808749516066589, | |
| "eval_loss": 0.41560646891593933, | |
| "eval_runtime": 145.9957, | |
| "eval_samples_per_second": 38.741, | |
| "eval_steps_per_second": 4.843, | |
| "eval_wer": 0.3300540835486511, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.800232288037166, | |
| "eval_loss": 0.41584905982017517, | |
| "eval_runtime": 147.0182, | |
| "eval_samples_per_second": 38.471, | |
| "eval_steps_per_second": 4.809, | |
| "eval_wer": 0.33032690857152025, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 1.8195896244676733, | |
| "eval_loss": 0.41545388102531433, | |
| "eval_runtime": 147.1819, | |
| "eval_samples_per_second": 38.429, | |
| "eval_steps_per_second": 4.804, | |
| "eval_wer": 0.32944423937988476, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.8389469608981805, | |
| "grad_norm": 0.9967782497406006, | |
| "learning_rate": 1.6073684210526313e-05, | |
| "loss": 0.37, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8389469608981805, | |
| "eval_loss": 0.41372692584991455, | |
| "eval_runtime": 146.2893, | |
| "eval_samples_per_second": 38.663, | |
| "eval_steps_per_second": 4.833, | |
| "eval_wer": 0.32921955994928664, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8583042973286876, | |
| "eval_loss": 0.4120025932788849, | |
| "eval_runtime": 146.1391, | |
| "eval_samples_per_second": 38.703, | |
| "eval_steps_per_second": 4.838, | |
| "eval_wer": 0.3284492304729502, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.8776616337591947, | |
| "eval_loss": 0.4108966886997223, | |
| "eval_runtime": 146.9334, | |
| "eval_samples_per_second": 38.494, | |
| "eval_steps_per_second": 4.812, | |
| "eval_wer": 0.3300701320794081, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 1.897018970189702, | |
| "eval_loss": 0.4100329577922821, | |
| "eval_runtime": 146.8452, | |
| "eval_samples_per_second": 38.517, | |
| "eval_steps_per_second": 4.815, | |
| "eval_wer": 0.32785543483494084, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.916376306620209, | |
| "eval_loss": 0.4094770848751068, | |
| "eval_runtime": 146.5252, | |
| "eval_samples_per_second": 38.601, | |
| "eval_steps_per_second": 4.825, | |
| "eval_wer": 0.3266999406204362, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "grad_norm": 0.7779282927513123, | |
| "learning_rate": 2.842105263157894e-07, | |
| "loss": 0.371, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "eval_loss": 0.409473717212677, | |
| "eval_runtime": 146.0247, | |
| "eval_samples_per_second": 38.733, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.3270690568278474, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "step": 10000, | |
| "total_flos": 1.1255918428180738e+19, | |
| "train_loss": 0.7365739318847656, | |
| "train_runtime": 19473.5524, | |
| "train_samples_per_second": 4.108, | |
| "train_steps_per_second": 0.514 | |
| } | |
| ], | |
| "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 | |
| } | |