| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7858854964831624, | |
| "eval_steps": 200, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007858854964831625, | |
| "eval_loss": 3.1856138706207275, | |
| "eval_runtime": 145.9906, | |
| "eval_samples_per_second": 38.742, | |
| "eval_steps_per_second": 4.843, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.01571770992966325, | |
| "eval_loss": 2.649242877960205, | |
| "eval_runtime": 145.4142, | |
| "eval_samples_per_second": 38.896, | |
| "eval_steps_per_second": 4.862, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.01964713741207906, | |
| "grad_norm": 2.1910934448242188, | |
| "learning_rate": 0.0002982, | |
| "loss": 4.6997, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.023576564894494872, | |
| "eval_loss": 1.3868569135665894, | |
| "eval_runtime": 144.4791, | |
| "eval_samples_per_second": 39.148, | |
| "eval_steps_per_second": 4.893, | |
| "eval_wer": 0.8722215981126927, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.0314354198593265, | |
| "eval_loss": 1.230230689048767, | |
| "eval_runtime": 145.2382, | |
| "eval_samples_per_second": 38.943, | |
| "eval_steps_per_second": 4.868, | |
| "eval_wer": 0.830768243167338, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.03929427482415812, | |
| "grad_norm": 4.400168418884277, | |
| "learning_rate": 0.0002923538461538461, | |
| "loss": 1.0569, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03929427482415812, | |
| "eval_loss": 1.1379698514938354, | |
| "eval_runtime": 146.1828, | |
| "eval_samples_per_second": 38.691, | |
| "eval_steps_per_second": 4.836, | |
| "eval_wer": 0.795846640240086, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.047153129788989744, | |
| "eval_loss": 1.066832184791565, | |
| "eval_runtime": 145.7914, | |
| "eval_samples_per_second": 38.795, | |
| "eval_steps_per_second": 4.849, | |
| "eval_wer": 0.7697998748214601, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.055011984753821366, | |
| "eval_loss": 1.0207505226135254, | |
| "eval_runtime": 146.2293, | |
| "eval_samples_per_second": 38.679, | |
| "eval_steps_per_second": 4.835, | |
| "eval_wer": 0.7310426730432829, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.05894141223623718, | |
| "grad_norm": 3.929137945175171, | |
| "learning_rate": 0.00028466153846153845, | |
| "loss": 0.8131, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.062870839718653, | |
| "eval_loss": 0.970230758190155, | |
| "eval_runtime": 145.9066, | |
| "eval_samples_per_second": 38.765, | |
| "eval_steps_per_second": 4.846, | |
| "eval_wer": 0.7151385790630868, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.07072969468348461, | |
| "eval_loss": 0.9408352375030518, | |
| "eval_runtime": 146.5767, | |
| "eval_samples_per_second": 38.587, | |
| "eval_steps_per_second": 4.823, | |
| "eval_wer": 0.6882091444528253, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.07858854964831624, | |
| "grad_norm": 2.3885908126831055, | |
| "learning_rate": 0.0002769692307692307, | |
| "loss": 0.7194, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07858854964831624, | |
| "eval_loss": 0.9249575138092041, | |
| "eval_runtime": 145.9151, | |
| "eval_samples_per_second": 38.762, | |
| "eval_steps_per_second": 4.845, | |
| "eval_wer": 0.6804095585049189, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08644740461314787, | |
| "eval_loss": 0.9052397608757019, | |
| "eval_runtime": 146.242, | |
| "eval_samples_per_second": 38.676, | |
| "eval_steps_per_second": 4.834, | |
| "eval_wer": 0.6726099725570124, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.09430625957797949, | |
| "eval_loss": 0.8985734581947327, | |
| "eval_runtime": 146.806, | |
| "eval_samples_per_second": 38.527, | |
| "eval_steps_per_second": 4.816, | |
| "eval_wer": 0.6573478198070967, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.0982356870603953, | |
| "grad_norm": 1.7855921983718872, | |
| "learning_rate": 0.00026927692307692305, | |
| "loss": 0.6688, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.10216511454281112, | |
| "eval_loss": 0.8814770579338074, | |
| "eval_runtime": 147.0906, | |
| "eval_samples_per_second": 38.452, | |
| "eval_steps_per_second": 4.807, | |
| "eval_wer": 0.6473495851454799, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.11002396950764273, | |
| "eval_loss": 0.858833372592926, | |
| "eval_runtime": 146.5037, | |
| "eval_samples_per_second": 38.607, | |
| "eval_steps_per_second": 4.826, | |
| "eval_wer": 0.6444608496092182, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.11788282447247436, | |
| "grad_norm": 2.42130184173584, | |
| "learning_rate": 0.00026158461538461537, | |
| "loss": 0.645, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.11788282447247436, | |
| "eval_loss": 0.875824511051178, | |
| "eval_runtime": 147.7065, | |
| "eval_samples_per_second": 38.292, | |
| "eval_steps_per_second": 4.787, | |
| "eval_wer": 0.6487458073213397, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.125741679437306, | |
| "eval_loss": 0.8724836707115173, | |
| "eval_runtime": 146.8488, | |
| "eval_samples_per_second": 38.516, | |
| "eval_steps_per_second": 4.814, | |
| "eval_wer": 0.6690953443212274, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.13360053440213762, | |
| "eval_loss": 0.8295639157295227, | |
| "eval_runtime": 147.6357, | |
| "eval_samples_per_second": 38.311, | |
| "eval_steps_per_second": 4.789, | |
| "eval_wer": 0.6298085410280689, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.1375299618845534, | |
| "grad_norm": 3.927525520324707, | |
| "learning_rate": 0.0002538923076923077, | |
| "loss": 0.6077, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.14145938936696922, | |
| "eval_loss": 0.8355618715286255, | |
| "eval_runtime": 147.0713, | |
| "eval_samples_per_second": 38.458, | |
| "eval_steps_per_second": 4.807, | |
| "eval_wer": 0.6551652196241434, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.14931824433180085, | |
| "eval_loss": 0.8262892961502075, | |
| "eval_runtime": 147.1539, | |
| "eval_samples_per_second": 38.436, | |
| "eval_steps_per_second": 4.804, | |
| "eval_wer": 0.6228595272102839, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.15717709929663248, | |
| "grad_norm": 2.224973201751709, | |
| "learning_rate": 0.00024619999999999997, | |
| "loss": 0.5983, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.15717709929663248, | |
| "eval_loss": 0.8710989356040955, | |
| "eval_runtime": 147.2188, | |
| "eval_samples_per_second": 38.419, | |
| "eval_steps_per_second": 4.802, | |
| "eval_wer": 0.6884659209449375, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1650359542614641, | |
| "eval_loss": 0.7836620807647705, | |
| "eval_runtime": 148.3617, | |
| "eval_samples_per_second": 38.123, | |
| "eval_steps_per_second": 4.765, | |
| "eval_wer": 0.5918216687262281, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.17289480922629574, | |
| "eval_loss": 0.8097087144851685, | |
| "eval_runtime": 147.4155, | |
| "eval_samples_per_second": 38.368, | |
| "eval_steps_per_second": 4.796, | |
| "eval_wer": 0.659755099420648, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.17682423670871153, | |
| "grad_norm": 1.8362923860549927, | |
| "learning_rate": 0.0002385230769230769, | |
| "loss": 0.5788, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.18075366419112734, | |
| "eval_loss": 0.77768874168396, | |
| "eval_runtime": 146.8579, | |
| "eval_samples_per_second": 38.513, | |
| "eval_steps_per_second": 4.814, | |
| "eval_wer": 0.5869268668453403, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.18861251915595897, | |
| "eval_loss": 0.7912825345993042, | |
| "eval_runtime": 146.176, | |
| "eval_samples_per_second": 38.693, | |
| "eval_steps_per_second": 4.837, | |
| "eval_wer": 0.5895588258894898, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.1964713741207906, | |
| "grad_norm": 3.315845489501953, | |
| "learning_rate": 0.0002308307692307692, | |
| "loss": 0.5501, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1964713741207906, | |
| "eval_loss": 0.7924312353134155, | |
| "eval_runtime": 146.7719, | |
| "eval_samples_per_second": 38.536, | |
| "eval_steps_per_second": 4.817, | |
| "eval_wer": 0.5899760876891721, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.20433022908562223, | |
| "eval_loss": 0.7602530717849731, | |
| "eval_runtime": 146.9845, | |
| "eval_samples_per_second": 38.48, | |
| "eval_steps_per_second": 4.81, | |
| "eval_wer": 0.5737189260323218, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.21218908405045384, | |
| "eval_loss": 0.7750186920166016, | |
| "eval_runtime": 146.5887, | |
| "eval_samples_per_second": 38.584, | |
| "eval_steps_per_second": 4.823, | |
| "eval_wer": 0.5931697453098169, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.21611851153286965, | |
| "grad_norm": 9.320504188537598, | |
| "learning_rate": 0.00022313846153846153, | |
| "loss": 0.5694, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.22004793901528547, | |
| "eval_loss": 0.7516711950302124, | |
| "eval_runtime": 146.9947, | |
| "eval_samples_per_second": 38.478, | |
| "eval_steps_per_second": 4.81, | |
| "eval_wer": 0.5711190640496863, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.2279067939801171, | |
| "eval_loss": 0.7651358842849731, | |
| "eval_runtime": 146.6177, | |
| "eval_samples_per_second": 38.577, | |
| "eval_steps_per_second": 4.822, | |
| "eval_wer": 0.5698191330583685, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.23576564894494872, | |
| "grad_norm": 2.727358102798462, | |
| "learning_rate": 0.00021544615384615383, | |
| "loss": 0.5424, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.23576564894494872, | |
| "eval_loss": 0.7547870874404907, | |
| "eval_runtime": 146.6389, | |
| "eval_samples_per_second": 38.571, | |
| "eval_steps_per_second": 4.821, | |
| "eval_wer": 0.5820481134952095, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24362450390978035, | |
| "eval_loss": 0.730515718460083, | |
| "eval_runtime": 146.763, | |
| "eval_samples_per_second": 38.538, | |
| "eval_steps_per_second": 4.817, | |
| "eval_wer": 0.5681019402673685, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.251483358874612, | |
| "eval_loss": 0.7314247488975525, | |
| "eval_runtime": 147.0063, | |
| "eval_samples_per_second": 38.475, | |
| "eval_steps_per_second": 4.809, | |
| "eval_wer": 0.5589221806743593, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.2554127863570278, | |
| "grad_norm": 3.2329583168029785, | |
| "learning_rate": 0.00020775384615384613, | |
| "loss": 0.521, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.2593422138394436, | |
| "eval_loss": 0.7227704524993896, | |
| "eval_runtime": 147.7519, | |
| "eval_samples_per_second": 38.28, | |
| "eval_steps_per_second": 4.785, | |
| "eval_wer": 0.565437884161705, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.26720106880427524, | |
| "eval_loss": 0.7350090146064758, | |
| "eval_runtime": 147.7712, | |
| "eval_samples_per_second": 38.275, | |
| "eval_steps_per_second": 4.784, | |
| "eval_wer": 0.5633194781017797, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.2750599237691068, | |
| "grad_norm": 3.2193281650543213, | |
| "learning_rate": 0.00020006153846153843, | |
| "loss": 0.5119, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.2750599237691068, | |
| "eval_loss": 0.7079117298126221, | |
| "eval_runtime": 146.6317, | |
| "eval_samples_per_second": 38.573, | |
| "eval_steps_per_second": 4.822, | |
| "eval_wer": 0.5346888992312754, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.28291877873393845, | |
| "eval_loss": 0.7105109691619873, | |
| "eval_runtime": 147.7789, | |
| "eval_samples_per_second": 38.273, | |
| "eval_steps_per_second": 4.784, | |
| "eval_wer": 0.5601097719503779, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.2907776336987701, | |
| "eval_loss": 0.6876121163368225, | |
| "eval_runtime": 147.4709, | |
| "eval_samples_per_second": 38.353, | |
| "eval_steps_per_second": 4.794, | |
| "eval_wer": 0.5378344112596491, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.2947070611811859, | |
| "grad_norm": 2.7452991008758545, | |
| "learning_rate": 0.00019236923076923075, | |
| "loss": 0.5007, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2986364886636017, | |
| "eval_loss": 0.6834765076637268, | |
| "eval_runtime": 147.74, | |
| "eval_samples_per_second": 38.283, | |
| "eval_steps_per_second": 4.785, | |
| "eval_wer": 0.5303397473961259, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.30649534362843334, | |
| "eval_loss": 0.7131712436676025, | |
| "eval_runtime": 147.6824, | |
| "eval_samples_per_second": 38.298, | |
| "eval_steps_per_second": 4.787, | |
| "eval_wer": 0.5350740639694436, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.31435419859326497, | |
| "grad_norm": 2.4165494441986084, | |
| "learning_rate": 0.00018467692307692308, | |
| "loss": 0.4934, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.31435419859326497, | |
| "eval_loss": 0.697209358215332, | |
| "eval_runtime": 146.9527, | |
| "eval_samples_per_second": 38.489, | |
| "eval_steps_per_second": 4.811, | |
| "eval_wer": 0.5241771115854343, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.3222130535580966, | |
| "eval_loss": 0.680029571056366, | |
| "eval_runtime": 147.261, | |
| "eval_samples_per_second": 38.408, | |
| "eval_steps_per_second": 4.801, | |
| "eval_wer": 0.5226685496942755, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.3300719085229282, | |
| "eval_loss": 0.6915732026100159, | |
| "eval_runtime": 146.7546, | |
| "eval_samples_per_second": 38.541, | |
| "eval_steps_per_second": 4.818, | |
| "eval_wer": 0.5364702861453033, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.33400133600534404, | |
| "grad_norm": 2.036782741546631, | |
| "learning_rate": 0.00017698461538461537, | |
| "loss": 0.4762, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.33793076348775986, | |
| "eval_loss": 0.6801823377609253, | |
| "eval_runtime": 147.6195, | |
| "eval_samples_per_second": 38.315, | |
| "eval_steps_per_second": 4.789, | |
| "eval_wer": 0.5255251881690232, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.3457896184525915, | |
| "eval_loss": 0.6977699398994446, | |
| "eval_runtime": 148.3269, | |
| "eval_samples_per_second": 38.132, | |
| "eval_steps_per_second": 4.766, | |
| "eval_wer": 0.5336938903243408, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.35364847341742306, | |
| "grad_norm": 3.6703684329986572, | |
| "learning_rate": 0.0001693076923076923, | |
| "loss": 0.4774, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.35364847341742306, | |
| "eval_loss": 0.6566863059997559, | |
| "eval_runtime": 147.4913, | |
| "eval_samples_per_second": 38.348, | |
| "eval_steps_per_second": 4.794, | |
| "eval_wer": 0.5210957936800886, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.3615073283822547, | |
| "eval_loss": 0.6478887796401978, | |
| "eval_runtime": 146.7975, | |
| "eval_samples_per_second": 38.529, | |
| "eval_steps_per_second": 4.816, | |
| "eval_wer": 0.5152380799537802, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.3693661833470863, | |
| "eval_loss": 0.6551229953765869, | |
| "eval_runtime": 147.2799, | |
| "eval_samples_per_second": 38.403, | |
| "eval_steps_per_second": 4.8, | |
| "eval_wer": 0.5147405755003129, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.37329561082950213, | |
| "grad_norm": 2.4989895820617676, | |
| "learning_rate": 0.00016161538461538462, | |
| "loss": 0.4632, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.37722503831191795, | |
| "eval_loss": 0.6358110308647156, | |
| "eval_runtime": 148.0942, | |
| "eval_samples_per_second": 38.192, | |
| "eval_steps_per_second": 4.774, | |
| "eval_wer": 0.4954502415303879, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.3850838932767496, | |
| "eval_loss": 0.6466320157051086, | |
| "eval_runtime": 147.5131, | |
| "eval_samples_per_second": 38.342, | |
| "eval_steps_per_second": 4.793, | |
| "eval_wer": 0.5109049766493877, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.3929427482415812, | |
| "grad_norm": 1.681718349456787, | |
| "learning_rate": 0.00015392307692307691, | |
| "loss": 0.4483, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.3929427482415812, | |
| "eval_loss": 0.6306164264678955, | |
| "eval_runtime": 147.9882, | |
| "eval_samples_per_second": 38.219, | |
| "eval_steps_per_second": 4.777, | |
| "eval_wer": 0.504421370223556, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.40080160320641284, | |
| "eval_loss": 0.6359797716140747, | |
| "eval_runtime": 147.568, | |
| "eval_samples_per_second": 38.328, | |
| "eval_steps_per_second": 4.791, | |
| "eval_wer": 0.5003771404727897, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.40866045817124447, | |
| "eval_loss": 0.6301611661911011, | |
| "eval_runtime": 146.6632, | |
| "eval_samples_per_second": 38.565, | |
| "eval_steps_per_second": 4.821, | |
| "eval_wer": 0.49135786618735056, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.4125898856536603, | |
| "grad_norm": 2.8097307682037354, | |
| "learning_rate": 0.0001462307692307692, | |
| "loss": 0.4454, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.4165193131360761, | |
| "eval_loss": 0.616253674030304, | |
| "eval_runtime": 147.6301, | |
| "eval_samples_per_second": 38.312, | |
| "eval_steps_per_second": 4.789, | |
| "eval_wer": 0.48505079359984593, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.42437816810090767, | |
| "eval_loss": 0.6221349239349365, | |
| "eval_runtime": 147.7897, | |
| "eval_samples_per_second": 38.271, | |
| "eval_steps_per_second": 4.784, | |
| "eval_wer": 0.49105294410296735, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.4322370230657393, | |
| "grad_norm": 4.2350687980651855, | |
| "learning_rate": 0.0001385384615384615, | |
| "loss": 0.4302, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.4322370230657393, | |
| "eval_loss": 0.6395624279975891, | |
| "eval_runtime": 147.7882, | |
| "eval_samples_per_second": 38.271, | |
| "eval_steps_per_second": 4.784, | |
| "eval_wer": 0.5000561698576496, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.44009587803057093, | |
| "eval_loss": 0.6212363839149475, | |
| "eval_runtime": 148.3281, | |
| "eval_samples_per_second": 38.132, | |
| "eval_steps_per_second": 4.766, | |
| "eval_wer": 0.4840557846929114, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.44795473299540256, | |
| "eval_loss": 0.6267797946929932, | |
| "eval_runtime": 147.5749, | |
| "eval_samples_per_second": 38.326, | |
| "eval_steps_per_second": 4.791, | |
| "eval_wer": 0.49379724286241594, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.4518841604778184, | |
| "grad_norm": 2.2473807334899902, | |
| "learning_rate": 0.00013086153846153845, | |
| "loss": 0.4261, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.4558135879602342, | |
| "eval_loss": 0.6097697019577026, | |
| "eval_runtime": 148.2001, | |
| "eval_samples_per_second": 38.165, | |
| "eval_steps_per_second": 4.771, | |
| "eval_wer": 0.48201762128677117, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.4636724429250658, | |
| "eval_loss": 0.6009463667869568, | |
| "eval_runtime": 147.4302, | |
| "eval_samples_per_second": 38.364, | |
| "eval_steps_per_second": 4.795, | |
| "eval_wer": 0.4689220201890517, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.47153129788989745, | |
| "grad_norm": 2.0571179389953613, | |
| "learning_rate": 0.00012316923076923078, | |
| "loss": 0.4026, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.47153129788989745, | |
| "eval_loss": 0.609122633934021, | |
| "eval_runtime": 147.8463, | |
| "eval_samples_per_second": 38.256, | |
| "eval_steps_per_second": 4.782, | |
| "eval_wer": 0.4809584182568086, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.4793901528547291, | |
| "eval_loss": 0.6019255518913269, | |
| "eval_runtime": 148.5912, | |
| "eval_samples_per_second": 38.064, | |
| "eval_steps_per_second": 4.758, | |
| "eval_wer": 0.4805732535186404, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 0.4872490078195607, | |
| "eval_loss": 0.5946715474128723, | |
| "eval_runtime": 147.4021, | |
| "eval_samples_per_second": 38.371, | |
| "eval_steps_per_second": 4.796, | |
| "eval_wer": 0.4671085362135097, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 0.4911784353019765, | |
| "grad_norm": 2.5033822059631348, | |
| "learning_rate": 0.00011547692307692306, | |
| "loss": 0.4027, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.49510786278439234, | |
| "eval_loss": 0.5993836522102356, | |
| "eval_runtime": 147.7878, | |
| "eval_samples_per_second": 38.271, | |
| "eval_steps_per_second": 4.784, | |
| "eval_wer": 0.47092808653367785, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 0.502966717749224, | |
| "eval_loss": 0.5981957912445068, | |
| "eval_runtime": 147.5137, | |
| "eval_samples_per_second": 38.342, | |
| "eval_steps_per_second": 4.793, | |
| "eval_wer": 0.4760796649066778, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 0.5108255727140556, | |
| "grad_norm": 3.0013859272003174, | |
| "learning_rate": 0.00010778461538461537, | |
| "loss": 0.3978, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.5108255727140556, | |
| "eval_loss": 0.5889731645584106, | |
| "eval_runtime": 148.3841, | |
| "eval_samples_per_second": 38.117, | |
| "eval_steps_per_second": 4.765, | |
| "eval_wer": 0.4632087432395564, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.5186844276788872, | |
| "eval_loss": 0.5871375799179077, | |
| "eval_runtime": 147.6289, | |
| "eval_samples_per_second": 38.312, | |
| "eval_steps_per_second": 4.789, | |
| "eval_wer": 0.4567090882829677, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 0.5265432826437189, | |
| "eval_loss": 0.5873442888259888, | |
| "eval_runtime": 148.2022, | |
| "eval_samples_per_second": 38.164, | |
| "eval_steps_per_second": 4.771, | |
| "eval_wer": 0.4634815682624256, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 0.5304727101261346, | |
| "grad_norm": 3.4298863410949707, | |
| "learning_rate": 0.00010009230769230768, | |
| "loss": 0.3875, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.5344021376085505, | |
| "eval_loss": 0.5772218704223633, | |
| "eval_runtime": 148.5655, | |
| "eval_samples_per_second": 38.071, | |
| "eval_steps_per_second": 4.759, | |
| "eval_wer": 0.4538684983389771, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 0.542260992573382, | |
| "eval_loss": 0.5603720545768738, | |
| "eval_runtime": 147.4995, | |
| "eval_samples_per_second": 38.346, | |
| "eval_steps_per_second": 4.793, | |
| "eval_wer": 0.4419123429250052, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 0.5501198475382136, | |
| "grad_norm": 2.6594979763031006, | |
| "learning_rate": 9.24e-05, | |
| "loss": 0.404, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.5501198475382136, | |
| "eval_loss": 0.5688683986663818, | |
| "eval_runtime": 147.0269, | |
| "eval_samples_per_second": 38.469, | |
| "eval_steps_per_second": 4.809, | |
| "eval_wer": 0.4454269711607902, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.5579787025030453, | |
| "eval_loss": 0.5594531893730164, | |
| "eval_runtime": 147.2136, | |
| "eval_samples_per_second": 38.42, | |
| "eval_steps_per_second": 4.803, | |
| "eval_wer": 0.443292516570108, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 0.5658375574678769, | |
| "eval_loss": 0.5574955940246582, | |
| "eval_runtime": 147.8377, | |
| "eval_samples_per_second": 38.258, | |
| "eval_steps_per_second": 4.782, | |
| "eval_wer": 0.4405803148721735, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 0.5697669849502928, | |
| "grad_norm": 1.834619402885437, | |
| "learning_rate": 8.472307692307691e-05, | |
| "loss": 0.3878, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.5736964124327085, | |
| "eval_loss": 0.5521669983863831, | |
| "eval_runtime": 147.831, | |
| "eval_samples_per_second": 38.26, | |
| "eval_steps_per_second": 4.782, | |
| "eval_wer": 0.43528429972236043, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 0.5815552673975402, | |
| "eval_loss": 0.5521777272224426, | |
| "eval_runtime": 148.1309, | |
| "eval_samples_per_second": 38.182, | |
| "eval_steps_per_second": 4.773, | |
| "eval_wer": 0.4351719600070614, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 0.5894141223623718, | |
| "grad_norm": 2.739065408706665, | |
| "learning_rate": 7.703076923076922e-05, | |
| "loss": 0.3622, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5894141223623718, | |
| "eval_loss": 0.5570353865623474, | |
| "eval_runtime": 147.6686, | |
| "eval_samples_per_second": 38.302, | |
| "eval_steps_per_second": 4.788, | |
| "eval_wer": 0.4401309560109772, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.5972729773272034, | |
| "eval_loss": 0.5467315912246704, | |
| "eval_runtime": 147.8113, | |
| "eval_samples_per_second": 38.265, | |
| "eval_steps_per_second": 4.783, | |
| "eval_wer": 0.42803036382019227, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 0.605131832292035, | |
| "eval_loss": 0.5510929822921753, | |
| "eval_runtime": 148.0412, | |
| "eval_samples_per_second": 38.206, | |
| "eval_steps_per_second": 4.776, | |
| "eval_wer": 0.43404856285407073, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 0.6090612597744509, | |
| "grad_norm": 3.7551369667053223, | |
| "learning_rate": 6.933846153846154e-05, | |
| "loss": 0.3545, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.6129906872568667, | |
| "eval_loss": 0.5437116026878357, | |
| "eval_runtime": 147.4367, | |
| "eval_samples_per_second": 38.362, | |
| "eval_steps_per_second": 4.795, | |
| "eval_wer": 0.4245478326459213, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 0.6208495422216983, | |
| "eval_loss": 0.5488719940185547, | |
| "eval_runtime": 147.4687, | |
| "eval_samples_per_second": 38.354, | |
| "eval_steps_per_second": 4.794, | |
| "eval_wer": 0.4296352168958932, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 0.6287083971865299, | |
| "grad_norm": 1.4097563028335571, | |
| "learning_rate": 6.164615384615383e-05, | |
| "loss": 0.3486, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6287083971865299, | |
| "eval_loss": 0.5420017242431641, | |
| "eval_runtime": 147.3988, | |
| "eval_samples_per_second": 38.372, | |
| "eval_steps_per_second": 4.797, | |
| "eval_wer": 0.4277896358588371, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6365672521513616, | |
| "eval_loss": 0.5352106094360352, | |
| "eval_runtime": 148.3938, | |
| "eval_samples_per_second": 38.115, | |
| "eval_steps_per_second": 4.764, | |
| "eval_wer": 0.4212739323714914, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 0.6444261071161932, | |
| "eval_loss": 0.5376533269882202, | |
| "eval_runtime": 147.9443, | |
| "eval_samples_per_second": 38.231, | |
| "eval_steps_per_second": 4.779, | |
| "eval_wer": 0.42592800629102406, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 0.648355534598609, | |
| "grad_norm": 2.8873980045318604, | |
| "learning_rate": 5.395384615384615e-05, | |
| "loss": 0.3374, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.6522849620810248, | |
| "eval_loss": 0.5336429476737976, | |
| "eval_runtime": 149.7227, | |
| "eval_samples_per_second": 37.776, | |
| "eval_steps_per_second": 4.722, | |
| "eval_wer": 0.4305018375567717, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 0.6601438170458565, | |
| "eval_loss": 0.5293694734573364, | |
| "eval_runtime": 148.8049, | |
| "eval_samples_per_second": 38.01, | |
| "eval_steps_per_second": 4.751, | |
| "eval_wer": 0.41875431304264094, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 0.6680026720106881, | |
| "grad_norm": 3.3759422302246094, | |
| "learning_rate": 4.6276923076923074e-05, | |
| "loss": 0.3389, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.6680026720106881, | |
| "eval_loss": 0.5252575278282166, | |
| "eval_runtime": 149.4186, | |
| "eval_samples_per_second": 37.853, | |
| "eval_steps_per_second": 4.732, | |
| "eval_wer": 0.4169408290670989, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.6758615269755197, | |
| "eval_loss": 0.5194066166877747, | |
| "eval_runtime": 147.8703, | |
| "eval_samples_per_second": 38.25, | |
| "eval_steps_per_second": 4.781, | |
| "eval_wer": 0.41443725826900546, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 0.6837203819403513, | |
| "eval_loss": 0.5231760740280151, | |
| "eval_runtime": 147.3205, | |
| "eval_samples_per_second": 38.392, | |
| "eval_steps_per_second": 4.799, | |
| "eval_wer": 0.417101314374669, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 0.6876498094227671, | |
| "grad_norm": 3.4031200408935547, | |
| "learning_rate": 3.858461538461538e-05, | |
| "loss": 0.3258, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.691579236905183, | |
| "eval_loss": 0.5179495811462402, | |
| "eval_runtime": 148.0293, | |
| "eval_samples_per_second": 38.209, | |
| "eval_steps_per_second": 4.776, | |
| "eval_wer": 0.41653961579817367, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 0.6994380918700145, | |
| "eval_loss": 0.5132375359535217, | |
| "eval_runtime": 149.1897, | |
| "eval_samples_per_second": 37.911, | |
| "eval_steps_per_second": 4.739, | |
| "eval_wer": 0.4103769799874821, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 0.7072969468348461, | |
| "grad_norm": 4.0969133377075195, | |
| "learning_rate": 3.090769230769231e-05, | |
| "loss": 0.327, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.7072969468348461, | |
| "eval_loss": 0.5096033215522766, | |
| "eval_runtime": 148.384, | |
| "eval_samples_per_second": 38.117, | |
| "eval_steps_per_second": 4.765, | |
| "eval_wer": 0.40440692654587473, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.7151558017996777, | |
| "eval_loss": 0.5040988922119141, | |
| "eval_runtime": 148.3441, | |
| "eval_samples_per_second": 38.128, | |
| "eval_steps_per_second": 4.766, | |
| "eval_wer": 0.40337982057742616, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 0.7230146567645094, | |
| "eval_loss": 0.5013387203216553, | |
| "eval_runtime": 147.8316, | |
| "eval_samples_per_second": 38.26, | |
| "eval_steps_per_second": 4.782, | |
| "eval_wer": 0.3980677568968561, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 0.7269440842469252, | |
| "grad_norm": 2.58701491355896, | |
| "learning_rate": 2.3215384615384613e-05, | |
| "loss": 0.316, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.730873511729341, | |
| "eval_loss": 0.5074244737625122, | |
| "eval_runtime": 148.936, | |
| "eval_samples_per_second": 37.976, | |
| "eval_steps_per_second": 4.747, | |
| "eval_wer": 0.4064611384827719, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 0.7387323666941726, | |
| "eval_loss": 0.5014389753341675, | |
| "eval_runtime": 148.0107, | |
| "eval_samples_per_second": 38.213, | |
| "eval_steps_per_second": 4.777, | |
| "eval_wer": 0.40545008104508035, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 0.7465912216590043, | |
| "grad_norm": 1.975710391998291, | |
| "learning_rate": 1.552307692307692e-05, | |
| "loss": 0.3162, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.7465912216590043, | |
| "eval_loss": 0.4958648383617401, | |
| "eval_runtime": 148.7492, | |
| "eval_samples_per_second": 38.024, | |
| "eval_steps_per_second": 4.753, | |
| "eval_wer": 0.3998170467493701, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.7544500766238359, | |
| "eval_loss": 0.4930271506309509, | |
| "eval_runtime": 147.4379, | |
| "eval_samples_per_second": 38.362, | |
| "eval_steps_per_second": 4.795, | |
| "eval_wer": 0.39818009661215514, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 0.7623089315886675, | |
| "eval_loss": 0.4924609363079071, | |
| "eval_runtime": 149.2355, | |
| "eval_samples_per_second": 37.9, | |
| "eval_steps_per_second": 4.737, | |
| "eval_wer": 0.39821219367366917, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 0.7662383590710834, | |
| "grad_norm": 2.332016944885254, | |
| "learning_rate": 7.83076923076923e-06, | |
| "loss": 0.3145, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.7701677865534992, | |
| "eval_loss": 0.4922255575656891, | |
| "eval_runtime": 148.1982, | |
| "eval_samples_per_second": 38.165, | |
| "eval_steps_per_second": 4.771, | |
| "eval_wer": 0.3970246023976505, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 0.7780266415183308, | |
| "eval_loss": 0.49084585905075073, | |
| "eval_runtime": 148.8021, | |
| "eval_samples_per_second": 38.01, | |
| "eval_steps_per_second": 4.751, | |
| "eval_wer": 0.39692831121310845, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 0.7858854964831624, | |
| "grad_norm": 5.262033462524414, | |
| "learning_rate": 1.5384615384615385e-07, | |
| "loss": 0.3095, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.7858854964831624, | |
| "eval_loss": 0.4908364713191986, | |
| "eval_runtime": 148.1385, | |
| "eval_samples_per_second": 38.18, | |
| "eval_steps_per_second": 4.773, | |
| "eval_wer": 0.39638266116737014, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.7858854964831624, | |
| "step": 20000, | |
| "total_flos": 2.4863424513490096e+19, | |
| "train_loss": 0.5819183097839356, | |
| "train_runtime": 23325.1985, | |
| "train_samples_per_second": 6.86, | |
| "train_steps_per_second": 0.857 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 2.4863424513490096e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |