| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0726161106939827, | |
| "eval_steps": 200, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.021452322213879653, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 109.0854, | |
| "eval_samples_per_second": 35.761, | |
| "eval_steps_per_second": 0.559, | |
| "eval_wer": 0.5591701685746027, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.042904644427759306, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.5961, | |
| "eval_samples_per_second": 36.943, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.4289474955320485, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.05363080553469913, | |
| "grad_norm": 2.6033225059509277, | |
| "learning_rate": 0.0001494, | |
| "loss": 2.1964, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06435696664163895, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.8581, | |
| "eval_samples_per_second": 37.203, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.43744867893542, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.08580928885551861, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.5471, | |
| "eval_samples_per_second": 37.313, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.4944452494807516, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.10726161106939826, | |
| "grad_norm": 4.917770862579346, | |
| "learning_rate": 0.0002988, | |
| "loss": 0.8327, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10726161106939826, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.2475, | |
| "eval_samples_per_second": 37.421, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.5149736753127566, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1287139332832779, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.0588, | |
| "eval_samples_per_second": 37.488, | |
| "eval_steps_per_second": 0.586, | |
| "eval_wer": 0.5633966091870743, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.15016625549715756, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.6247, | |
| "eval_samples_per_second": 37.286, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.5355021011447616, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.1608924166040974, | |
| "grad_norm": 2.821734666824341, | |
| "learning_rate": 0.00028346666666666665, | |
| "loss": 0.91, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.17161857771103722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.8778, | |
| "eval_samples_per_second": 37.196, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.515239337294112, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.19307089992491688, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.4509, | |
| "eval_samples_per_second": 36.994, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.5594599816451722, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.21452322213879652, | |
| "grad_norm": 9.015162467956543, | |
| "learning_rate": 0.0002668, | |
| "loss": 0.8721, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21452322213879652, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.3779, | |
| "eval_samples_per_second": 37.019, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.5056513548761049, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23597554435267618, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.6302, | |
| "eval_samples_per_second": 36.931, | |
| "eval_steps_per_second": 0.577, | |
| "eval_wer": 0.5041298362556151, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.2574278665665558, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.5892, | |
| "eval_samples_per_second": 36.945, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.5145631067961165, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.26815402767349567, | |
| "grad_norm": 5.016167163848877, | |
| "learning_rate": 0.0002501333333333333, | |
| "loss": 0.8218, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27888018878043547, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 106.4951, | |
| "eval_samples_per_second": 36.631, | |
| "eval_steps_per_second": 0.573, | |
| "eval_wer": 0.5018113316910593, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.3003325109943151, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 106.2902, | |
| "eval_samples_per_second": 36.701, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.5090566584552964, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.3217848332081948, | |
| "grad_norm": 2.5943267345428467, | |
| "learning_rate": 0.00023346666666666666, | |
| "loss": 0.8469, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3217848332081948, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.8685, | |
| "eval_samples_per_second": 36.848, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.5036709655605468, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.34323715542207445, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.9311, | |
| "eval_samples_per_second": 36.826, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.4703183113558421, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.3646894776359541, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.4279, | |
| "eval_samples_per_second": 37.002, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.47951987634642323, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.3754156387428939, | |
| "grad_norm": 4.555402755737305, | |
| "learning_rate": 0.0002168333333333333, | |
| "loss": 0.8142, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.38614179984983377, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.6085, | |
| "eval_samples_per_second": 36.938, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.4714051103704777, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.40759412206371337, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.5848, | |
| "eval_samples_per_second": 36.947, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.4553929382215138, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.42904644427759303, | |
| "grad_norm": 15.551188468933105, | |
| "learning_rate": 0.0002002, | |
| "loss": 0.8085, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42904644427759303, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.8874, | |
| "eval_samples_per_second": 36.841, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.4505868714679032, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4504987664914727, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 107.6738, | |
| "eval_samples_per_second": 36.23, | |
| "eval_steps_per_second": 0.567, | |
| "eval_wer": 0.4457566536250785, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.47195108870535235, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.5778, | |
| "eval_samples_per_second": 36.949, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.43669999516978214, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.4826772498122922, | |
| "grad_norm": 4.841684818267822, | |
| "learning_rate": 0.0001835333333333333, | |
| "loss": 0.7802, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.493403410919232, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.9573, | |
| "eval_samples_per_second": 36.817, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.4401052987489736, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.5148557331331116, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.3523, | |
| "eval_samples_per_second": 37.028, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.43336714485823313, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5363080553469913, | |
| "grad_norm": 7.372885227203369, | |
| "learning_rate": 0.0001669, | |
| "loss": 0.7493, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5363080553469913, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 107.6356, | |
| "eval_samples_per_second": 36.243, | |
| "eval_steps_per_second": 0.567, | |
| "eval_wer": 0.4224267014442351, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5577603775608709, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 107.4854, | |
| "eval_samples_per_second": 36.293, | |
| "eval_steps_per_second": 0.568, | |
| "eval_wer": 0.43278751871709414, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.5792126997747507, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.275, | |
| "eval_samples_per_second": 37.055, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.41764478577983866, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.5899388608816905, | |
| "grad_norm": 2.7961230278015137, | |
| "learning_rate": 0.00015023333333333332, | |
| "loss": 0.7668, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6006650219886303, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.464, | |
| "eval_samples_per_second": 36.989, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.41829686518862, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.62211734420251, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.96, | |
| "eval_samples_per_second": 37.167, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.40296092353765156, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6435696664163896, | |
| "grad_norm": 6.007960319519043, | |
| "learning_rate": 0.0001336, | |
| "loss": 0.6999, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6435696664163896, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.9116, | |
| "eval_samples_per_second": 37.184, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.4124523015988021, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6650219886302692, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 108.5507, | |
| "eval_samples_per_second": 35.937, | |
| "eval_steps_per_second": 0.562, | |
| "eval_wer": 0.40759793266676325, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.6864743108441489, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.9858, | |
| "eval_samples_per_second": 37.157, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.39170651596386996, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.6972004719510887, | |
| "grad_norm": 44.30250549316406, | |
| "learning_rate": 0.00011693333333333332, | |
| "loss": 0.6918, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7079266330580285, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 106.5414, | |
| "eval_samples_per_second": 36.615, | |
| "eval_steps_per_second": 0.573, | |
| "eval_wer": 0.4004250591701686, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.7293789552719082, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.9171, | |
| "eval_samples_per_second": 37.182, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.38653818287204755, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7508312774857878, | |
| "grad_norm": 3.788344144821167, | |
| "learning_rate": 0.00010029999999999998, | |
| "loss": 0.6888, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7508312774857878, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.3057, | |
| "eval_samples_per_second": 37.045, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.3785200212529585, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7722835996996675, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.7325, | |
| "eval_samples_per_second": 37.247, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.3824083466164324, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.7937359219135471, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.0488, | |
| "eval_samples_per_second": 37.135, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.37426942955127274, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.8044620830204869, | |
| "grad_norm": 5.486635684967041, | |
| "learning_rate": 8.363333333333332e-05, | |
| "loss": 0.646, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.8151882441274267, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 106.6481, | |
| "eval_samples_per_second": 36.578, | |
| "eval_steps_per_second": 0.572, | |
| "eval_wer": 0.3673139158576052, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.8366405663413065, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.8533, | |
| "eval_samples_per_second": 37.204, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.36668598753803794, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.8580928885551861, | |
| "grad_norm": 3.9184212684631348, | |
| "learning_rate": 6.696666666666666e-05, | |
| "loss": 0.6324, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8580928885551861, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.6572, | |
| "eval_samples_per_second": 36.921, | |
| "eval_steps_per_second": 0.577, | |
| "eval_wer": 0.3661546635753272, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8795452107690658, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.6274, | |
| "eval_samples_per_second": 37.285, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.36009274018258225, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.9009975329829454, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 104.9439, | |
| "eval_samples_per_second": 37.172, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.35345119064869823, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9117236940898852, | |
| "grad_norm": 3.5586395263671875, | |
| "learning_rate": 5.0299999999999996e-05, | |
| "loss": 0.6221, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9224498551968251, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.157, | |
| "eval_samples_per_second": 37.097, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.35258175143698983, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.9439021774107047, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.1867, | |
| "eval_samples_per_second": 37.086, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.34874172825194416, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.9653544996245844, | |
| "grad_norm": 3.914166212081909, | |
| "learning_rate": 3.363333333333333e-05, | |
| "loss": 0.6215, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9653544996245844, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.9335, | |
| "eval_samples_per_second": 36.825, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.34811379993237695, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.986806821838464, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.4283, | |
| "eval_samples_per_second": 37.001, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.3447084963531855, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.0082591440523436, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.2731, | |
| "eval_samples_per_second": 37.056, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.34103753079263877, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.0189853051592834, | |
| "grad_norm": 3.3699042797088623, | |
| "learning_rate": 1.6966666666666665e-05, | |
| "loss": 0.5603, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0297114662662232, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.4958, | |
| "eval_samples_per_second": 36.978, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.34053035791914216, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.051163788480103, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 105.0872, | |
| "eval_samples_per_second": 37.122, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.34120658841713764, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.0726161106939827, | |
| "grad_norm": 6.575745582580566, | |
| "learning_rate": 3.333333333333333e-07, | |
| "loss": 0.5284, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0726161106939827, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 108.9867, | |
| "eval_samples_per_second": 35.793, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 0.3401922426701444, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0726161106939827, | |
| "step": 10000, | |
| "total_flos": 4.496412338111517e+18, | |
| "train_loss": 0.8045109680175782, | |
| "train_runtime": 7510.9356, | |
| "train_samples_per_second": 5.326, | |
| "train_steps_per_second": 1.331 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.496412338111517e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |