| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.968210223964601, |
| "eval_steps": 2000, |
| "global_step": 64000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.015526142141831309, |
| "grad_norm": 337.8125, |
| "learning_rate": 1.4982145629560628e-07, |
| "loss": 33.1636, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.031052284283662617, |
| "grad_norm": 48.633304595947266, |
| "learning_rate": 3.04300574444962e-07, |
| "loss": 19.2812, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.046578426425493924, |
| "grad_norm": 21.59873390197754, |
| "learning_rate": 4.595559695699426e-07, |
| "loss": 11.9608, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.062104568567325234, |
| "grad_norm": 9.771416664123535, |
| "learning_rate": 6.148113646949232e-07, |
| "loss": 11.5474, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.07763071070915654, |
| "grad_norm": 23.054157257080078, |
| "learning_rate": 7.700667598199038e-07, |
| "loss": 11.1511, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.09315685285098785, |
| "grad_norm": 13.174666404724121, |
| "learning_rate": 9.253221549448845e-07, |
| "loss": 10.7244, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.10868299499281917, |
| "grad_norm": 1661.3245849609375, |
| "learning_rate": 1.080577550069865e-06, |
| "loss": 10.0404, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.12420913713465047, |
| "grad_norm": 15.175909996032715, |
| "learning_rate": 1.2358329451948456e-06, |
| "loss": 9.7509, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.13973527927648177, |
| "grad_norm": 28.184785842895508, |
| "learning_rate": 1.391088340319826e-06, |
| "loss": 9.5964, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.1552614214183131, |
| "grad_norm": 91.71954345703125, |
| "learning_rate": 1.546343735444807e-06, |
| "loss": 8.6393, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.1552614214183131, |
| "eval_loss": 4.543815612792969, |
| "eval_runtime": 1499.2099, |
| "eval_samples_per_second": 16.159, |
| "eval_steps_per_second": 2.02, |
| "eval_wer": 0.9787015626458175, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.1707875635601444, |
| "grad_norm": 25.240421295166016, |
| "learning_rate": 1.7015991305697874e-06, |
| "loss": 8.6948, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.1863137057019757, |
| "grad_norm": 14.96767520904541, |
| "learning_rate": 1.8568545256947683e-06, |
| "loss": 8.1006, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.201839847843807, |
| "grad_norm": 12.3013277053833, |
| "learning_rate": 2.0121099208197485e-06, |
| "loss": 9.8169, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.21736598998563833, |
| "grad_norm": 344.09686279296875, |
| "learning_rate": 2.167365315944729e-06, |
| "loss": 7.252, |
| "step": 2800 |
| }, |
| { |
| "epoch": 0.23289213212746962, |
| "grad_norm": 16.876171112060547, |
| "learning_rate": 2.32262071106971e-06, |
| "loss": 5.9625, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.24841827426930094, |
| "grad_norm": 8.259666442871094, |
| "learning_rate": 2.4778761061946905e-06, |
| "loss": 4.6505, |
| "step": 3200 |
| }, |
| { |
| "epoch": 0.26394441641113225, |
| "grad_norm": 7.110565662384033, |
| "learning_rate": 2.633131501319671e-06, |
| "loss": 4.4937, |
| "step": 3400 |
| }, |
| { |
| "epoch": 0.27947055855296354, |
| "grad_norm": 3.901717185974121, |
| "learning_rate": 2.7883868964446514e-06, |
| "loss": 3.3902, |
| "step": 3600 |
| }, |
| { |
| "epoch": 0.29499670069479483, |
| "grad_norm": 4.588987827301025, |
| "learning_rate": 2.943642291569632e-06, |
| "loss": 3.2178, |
| "step": 3800 |
| }, |
| { |
| "epoch": 0.3105228428366262, |
| "grad_norm": 12.945244789123535, |
| "learning_rate": 3.098897686694613e-06, |
| "loss": 2.1468, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.3105228428366262, |
| "eval_loss": 0.46712690591812134, |
| "eval_runtime": 1493.3754, |
| "eval_samples_per_second": 16.222, |
| "eval_steps_per_second": 2.028, |
| "eval_wer": 0.461719325169278, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.32604898497845747, |
| "grad_norm": 4.833748817443848, |
| "learning_rate": 3.254153081819594e-06, |
| "loss": 2.5828, |
| "step": 4200 |
| }, |
| { |
| "epoch": 0.3415751271202888, |
| "grad_norm": 2.971928596496582, |
| "learning_rate": 3.409408476944574e-06, |
| "loss": 2.3426, |
| "step": 4400 |
| }, |
| { |
| "epoch": 0.3571012692621201, |
| "grad_norm": 26.230440139770508, |
| "learning_rate": 3.5646638720695547e-06, |
| "loss": 2.488, |
| "step": 4600 |
| }, |
| { |
| "epoch": 0.3726274114039514, |
| "grad_norm": 6.009104251861572, |
| "learning_rate": 3.7199192671945354e-06, |
| "loss": 1.8542, |
| "step": 4800 |
| }, |
| { |
| "epoch": 0.38815355354578274, |
| "grad_norm": 6.4187421798706055, |
| "learning_rate": 3.875174662319516e-06, |
| "loss": 2.361, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.403679695687614, |
| "grad_norm": 6.448857307434082, |
| "learning_rate": 4.030430057444496e-06, |
| "loss": 2.0736, |
| "step": 5200 |
| }, |
| { |
| "epoch": 0.4192058378294453, |
| "grad_norm": 5.732870578765869, |
| "learning_rate": 4.185685452569477e-06, |
| "loss": 1.662, |
| "step": 5400 |
| }, |
| { |
| "epoch": 0.43473197997127666, |
| "grad_norm": 6.17863130569458, |
| "learning_rate": 4.340940847694458e-06, |
| "loss": 1.9072, |
| "step": 5600 |
| }, |
| { |
| "epoch": 0.45025812211310795, |
| "grad_norm": 15.475833892822266, |
| "learning_rate": 4.496196242819438e-06, |
| "loss": 1.857, |
| "step": 5800 |
| }, |
| { |
| "epoch": 0.46578426425493924, |
| "grad_norm": 7.531778812408447, |
| "learning_rate": 4.651451637944419e-06, |
| "loss": 1.8339, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.46578426425493924, |
| "eval_loss": 0.2383137196302414, |
| "eval_runtime": 1537.7831, |
| "eval_samples_per_second": 15.753, |
| "eval_steps_per_second": 1.97, |
| "eval_wer": 0.25908864670931886, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.4813104063967706, |
| "grad_norm": 7.888195037841797, |
| "learning_rate": 4.8067070330694e-06, |
| "loss": 1.6757, |
| "step": 6200 |
| }, |
| { |
| "epoch": 0.4968365485386019, |
| "grad_norm": 4.169632911682129, |
| "learning_rate": 4.96196242819438e-06, |
| "loss": 1.6712, |
| "step": 6400 |
| }, |
| { |
| "epoch": 0.5123626906804332, |
| "grad_norm": 2.824876070022583, |
| "learning_rate": 4.98697579740896e-06, |
| "loss": 1.4468, |
| "step": 6600 |
| }, |
| { |
| "epoch": 0.5278888328222645, |
| "grad_norm": 4.886559963226318, |
| "learning_rate": 4.969725197950629e-06, |
| "loss": 1.2449, |
| "step": 6800 |
| }, |
| { |
| "epoch": 0.5434149749640957, |
| "grad_norm": 5.823539733886719, |
| "learning_rate": 4.952474598492298e-06, |
| "loss": 1.2058, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.5589411171059271, |
| "grad_norm": 10.134882926940918, |
| "learning_rate": 4.935223999033967e-06, |
| "loss": 1.1973, |
| "step": 7200 |
| }, |
| { |
| "epoch": 0.5744672592477584, |
| "grad_norm": 2.770137310028076, |
| "learning_rate": 4.917973399575636e-06, |
| "loss": 1.3946, |
| "step": 7400 |
| }, |
| { |
| "epoch": 0.5899934013895897, |
| "grad_norm": 4.204318523406982, |
| "learning_rate": 4.900722800117305e-06, |
| "loss": 1.178, |
| "step": 7600 |
| }, |
| { |
| "epoch": 0.605519543531421, |
| "grad_norm": 17.23874282836914, |
| "learning_rate": 4.883472200658974e-06, |
| "loss": 1.1197, |
| "step": 7800 |
| }, |
| { |
| "epoch": 0.6210456856732524, |
| "grad_norm": 2.435584783554077, |
| "learning_rate": 4.866221601200642e-06, |
| "loss": 1.2017, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.6210456856732524, |
| "eval_loss": 0.17558206617832184, |
| "eval_runtime": 1514.9639, |
| "eval_samples_per_second": 15.99, |
| "eval_steps_per_second": 1.999, |
| "eval_wer": 0.18557067161625485, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.6365718278150837, |
| "grad_norm": 4.509705066680908, |
| "learning_rate": 4.848971001742311e-06, |
| "loss": 1.0055, |
| "step": 8200 |
| }, |
| { |
| "epoch": 0.6520979699569149, |
| "grad_norm": 3.2619400024414062, |
| "learning_rate": 4.83172040228398e-06, |
| "loss": 0.8313, |
| "step": 8400 |
| }, |
| { |
| "epoch": 0.6676241120987463, |
| "grad_norm": 3.4498465061187744, |
| "learning_rate": 4.8144698028256485e-06, |
| "loss": 1.0283, |
| "step": 8600 |
| }, |
| { |
| "epoch": 0.6831502542405776, |
| "grad_norm": 4.241556644439697, |
| "learning_rate": 4.797219203367317e-06, |
| "loss": 1.1568, |
| "step": 8800 |
| }, |
| { |
| "epoch": 0.6986763963824089, |
| "grad_norm": 3.5493125915527344, |
| "learning_rate": 4.779968603908986e-06, |
| "loss": 0.8577, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.7142025385242402, |
| "grad_norm": 35.144020080566406, |
| "learning_rate": 4.762718004450655e-06, |
| "loss": 0.9797, |
| "step": 9200 |
| }, |
| { |
| "epoch": 0.7297286806660715, |
| "grad_norm": 2.440404176712036, |
| "learning_rate": 4.745467404992324e-06, |
| "loss": 0.9456, |
| "step": 9400 |
| }, |
| { |
| "epoch": 0.7452548228079028, |
| "grad_norm": 7.8633599281311035, |
| "learning_rate": 4.728216805533993e-06, |
| "loss": 0.9581, |
| "step": 9600 |
| }, |
| { |
| "epoch": 0.7607809649497341, |
| "grad_norm": 10.246444702148438, |
| "learning_rate": 4.710966206075662e-06, |
| "loss": 0.8671, |
| "step": 9800 |
| }, |
| { |
| "epoch": 0.7763071070915655, |
| "grad_norm": 7.189419269561768, |
| "learning_rate": 4.693715606617331e-06, |
| "loss": 0.7, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.7763071070915655, |
| "eval_loss": 0.1570720672607422, |
| "eval_runtime": 1452.2458, |
| "eval_samples_per_second": 16.681, |
| "eval_steps_per_second": 2.086, |
| "eval_wer": 0.15577722705544436, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.7918332492333967, |
| "grad_norm": 2.7510902881622314, |
| "learning_rate": 4.676465007158999e-06, |
| "loss": 0.7451, |
| "step": 10200 |
| }, |
| { |
| "epoch": 0.807359391375228, |
| "grad_norm": 1.878658652305603, |
| "learning_rate": 4.659214407700668e-06, |
| "loss": 0.7284, |
| "step": 10400 |
| }, |
| { |
| "epoch": 0.8228855335170594, |
| "grad_norm": 4.4139275550842285, |
| "learning_rate": 4.642050061239628e-06, |
| "loss": 0.7892, |
| "step": 10600 |
| }, |
| { |
| "epoch": 0.8384116756588906, |
| "grad_norm": 2.93001127243042, |
| "learning_rate": 4.624799461781297e-06, |
| "loss": 0.7596, |
| "step": 10800 |
| }, |
| { |
| "epoch": 0.853937817800722, |
| "grad_norm": 5.795602798461914, |
| "learning_rate": 4.607548862322966e-06, |
| "loss": 0.6929, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.8694639599425533, |
| "grad_norm": 3.554075241088867, |
| "learning_rate": 4.590298262864635e-06, |
| "loss": 0.6534, |
| "step": 11200 |
| }, |
| { |
| "epoch": 0.8849901020843846, |
| "grad_norm": 3.8044190406799316, |
| "learning_rate": 4.573047663406304e-06, |
| "loss": 0.7146, |
| "step": 11400 |
| }, |
| { |
| "epoch": 0.9005162442262159, |
| "grad_norm": 0.9183899760246277, |
| "learning_rate": 4.555797063947973e-06, |
| "loss": 0.6882, |
| "step": 11600 |
| }, |
| { |
| "epoch": 0.9160423863680472, |
| "grad_norm": 1.4586737155914307, |
| "learning_rate": 4.538546464489642e-06, |
| "loss": 0.6639, |
| "step": 11800 |
| }, |
| { |
| "epoch": 0.9315685285098785, |
| "grad_norm": 1.5792444944381714, |
| "learning_rate": 4.5212958650313105e-06, |
| "loss": 0.6052, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.9315685285098785, |
| "eval_loss": 0.1463317573070526, |
| "eval_runtime": 1562.7951, |
| "eval_samples_per_second": 15.501, |
| "eval_steps_per_second": 1.938, |
| "eval_wer": 0.14558166302014744, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.9470946706517098, |
| "grad_norm": 5.483816623687744, |
| "learning_rate": 4.504045265572979e-06, |
| "loss": 0.6127, |
| "step": 12200 |
| }, |
| { |
| "epoch": 0.9626208127935412, |
| "grad_norm": 3.2851834297180176, |
| "learning_rate": 4.4867946661146475e-06, |
| "loss": 0.6252, |
| "step": 12400 |
| }, |
| { |
| "epoch": 0.9781469549353724, |
| "grad_norm": 6.533777713775635, |
| "learning_rate": 4.469544066656316e-06, |
| "loss": 0.6708, |
| "step": 12600 |
| }, |
| { |
| "epoch": 0.9936730970772037, |
| "grad_norm": 3.3322649002075195, |
| "learning_rate": 4.452293467197985e-06, |
| "loss": 0.589, |
| "step": 12800 |
| }, |
| { |
| "epoch": 1.0091604238636804, |
| "grad_norm": 4.870580196380615, |
| "learning_rate": 4.435042867739654e-06, |
| "loss": 0.6737, |
| "step": 13000 |
| }, |
| { |
| "epoch": 1.0246865660055118, |
| "grad_norm": 3.0527658462524414, |
| "learning_rate": 4.417792268281323e-06, |
| "loss": 0.5764, |
| "step": 13200 |
| }, |
| { |
| "epoch": 1.0402127081473431, |
| "grad_norm": 3.7095119953155518, |
| "learning_rate": 4.400541668822992e-06, |
| "loss": 0.6397, |
| "step": 13400 |
| }, |
| { |
| "epoch": 1.0557388502891745, |
| "grad_norm": 4.6553826332092285, |
| "learning_rate": 4.383291069364661e-06, |
| "loss": 0.4774, |
| "step": 13600 |
| }, |
| { |
| "epoch": 1.0712649924310056, |
| "grad_norm": 1.2344917058944702, |
| "learning_rate": 4.366126722903621e-06, |
| "loss": 0.5862, |
| "step": 13800 |
| }, |
| { |
| "epoch": 1.086791134572837, |
| "grad_norm": 1.4103082418441772, |
| "learning_rate": 4.34887612344529e-06, |
| "loss": 0.5755, |
| "step": 14000 |
| }, |
| { |
| "epoch": 1.086791134572837, |
| "eval_loss": 0.13507573306560516, |
| "eval_runtime": 1518.3896, |
| "eval_samples_per_second": 15.954, |
| "eval_steps_per_second": 1.995, |
| "eval_wer": 0.13720849448874417, |
| "step": 14000 |
| }, |
| { |
| "epoch": 1.1023172767146683, |
| "grad_norm": 7.926211833953857, |
| "learning_rate": 4.331625523986959e-06, |
| "loss": 0.5719, |
| "step": 14200 |
| }, |
| { |
| "epoch": 1.1178434188564996, |
| "grad_norm": 17.994617462158203, |
| "learning_rate": 4.314374924528628e-06, |
| "loss": 0.5771, |
| "step": 14400 |
| }, |
| { |
| "epoch": 1.133369560998331, |
| "grad_norm": 2.412418842315674, |
| "learning_rate": 4.297124325070297e-06, |
| "loss": 0.5532, |
| "step": 14600 |
| }, |
| { |
| "epoch": 1.1488957031401623, |
| "grad_norm": 0.9119038581848145, |
| "learning_rate": 4.279873725611966e-06, |
| "loss": 0.4524, |
| "step": 14800 |
| }, |
| { |
| "epoch": 1.1644218452819937, |
| "grad_norm": 0.5025124549865723, |
| "learning_rate": 4.262623126153635e-06, |
| "loss": 0.5582, |
| "step": 15000 |
| }, |
| { |
| "epoch": 1.1799479874238248, |
| "grad_norm": 1.1475783586502075, |
| "learning_rate": 4.245372526695303e-06, |
| "loss": 0.4976, |
| "step": 15200 |
| }, |
| { |
| "epoch": 1.1954741295656561, |
| "grad_norm": 3.657947063446045, |
| "learning_rate": 4.228121927236972e-06, |
| "loss": 0.6228, |
| "step": 15400 |
| }, |
| { |
| "epoch": 1.2110002717074875, |
| "grad_norm": 0.8675907850265503, |
| "learning_rate": 4.210871327778641e-06, |
| "loss": 0.6044, |
| "step": 15600 |
| }, |
| { |
| "epoch": 1.2265264138493188, |
| "grad_norm": 8.001958847045898, |
| "learning_rate": 4.1936207283203095e-06, |
| "loss": 0.5116, |
| "step": 15800 |
| }, |
| { |
| "epoch": 1.2420525559911502, |
| "grad_norm": 2.2227742671966553, |
| "learning_rate": 4.176370128861978e-06, |
| "loss": 0.5438, |
| "step": 16000 |
| }, |
| { |
| "epoch": 1.2420525559911502, |
| "eval_loss": 0.12887345254421234, |
| "eval_runtime": 1570.4952, |
| "eval_samples_per_second": 15.425, |
| "eval_steps_per_second": 1.929, |
| "eval_wer": 0.13367776521946514, |
| "step": 16000 |
| }, |
| { |
| "epoch": 1.2575786981329813, |
| "grad_norm": 2.515972137451172, |
| "learning_rate": 4.159119529403647e-06, |
| "loss": 0.5489, |
| "step": 16200 |
| }, |
| { |
| "epoch": 1.2731048402748129, |
| "grad_norm": 4.5147881507873535, |
| "learning_rate": 4.141868929945315e-06, |
| "loss": 0.5196, |
| "step": 16400 |
| }, |
| { |
| "epoch": 1.288630982416644, |
| "grad_norm": 2.1061394214630127, |
| "learning_rate": 4.124618330486984e-06, |
| "loss": 0.4751, |
| "step": 16600 |
| }, |
| { |
| "epoch": 1.3041571245584753, |
| "grad_norm": 1.0235892534255981, |
| "learning_rate": 4.107367731028653e-06, |
| "loss": 0.4246, |
| "step": 16800 |
| }, |
| { |
| "epoch": 1.3196832667003067, |
| "grad_norm": 1.3415697813034058, |
| "learning_rate": 4.090117131570322e-06, |
| "loss": 0.5067, |
| "step": 17000 |
| }, |
| { |
| "epoch": 1.335209408842138, |
| "grad_norm": 5.631282806396484, |
| "learning_rate": 4.072866532111992e-06, |
| "loss": 0.4272, |
| "step": 17200 |
| }, |
| { |
| "epoch": 1.3507355509839694, |
| "grad_norm": 3.1014630794525146, |
| "learning_rate": 4.05561593265366e-06, |
| "loss": 0.4893, |
| "step": 17400 |
| }, |
| { |
| "epoch": 1.3662616931258005, |
| "grad_norm": 1.6911731958389282, |
| "learning_rate": 4.03845158619262e-06, |
| "loss": 0.5704, |
| "step": 17600 |
| }, |
| { |
| "epoch": 1.3817878352676318, |
| "grad_norm": 2.359447717666626, |
| "learning_rate": 4.021200986734289e-06, |
| "loss": 0.5164, |
| "step": 17800 |
| }, |
| { |
| "epoch": 1.3973139774094632, |
| "grad_norm": 0.6359444856643677, |
| "learning_rate": 4.00403664027325e-06, |
| "loss": 0.4774, |
| "step": 18000 |
| }, |
| { |
| "epoch": 1.3973139774094632, |
| "eval_loss": 0.12409843504428864, |
| "eval_runtime": 1590.3608, |
| "eval_samples_per_second": 15.232, |
| "eval_steps_per_second": 1.905, |
| "eval_wer": 0.12936415765406112, |
| "step": 18000 |
| }, |
| { |
| "epoch": 1.4128401195512945, |
| "grad_norm": 1.8320542573928833, |
| "learning_rate": 3.986786040814918e-06, |
| "loss": 0.4498, |
| "step": 18200 |
| }, |
| { |
| "epoch": 1.4283662616931259, |
| "grad_norm": 1.884871006011963, |
| "learning_rate": 3.969535441356587e-06, |
| "loss": 0.54, |
| "step": 18400 |
| }, |
| { |
| "epoch": 1.443892403834957, |
| "grad_norm": 3.8327672481536865, |
| "learning_rate": 3.952284841898256e-06, |
| "loss": 0.5155, |
| "step": 18600 |
| }, |
| { |
| "epoch": 1.4594185459767886, |
| "grad_norm": 0.961982786655426, |
| "learning_rate": 3.935034242439925e-06, |
| "loss": 0.5327, |
| "step": 18800 |
| }, |
| { |
| "epoch": 1.4749446881186197, |
| "grad_norm": 2.743013858795166, |
| "learning_rate": 3.917783642981594e-06, |
| "loss": 0.4562, |
| "step": 19000 |
| }, |
| { |
| "epoch": 1.490470830260451, |
| "grad_norm": 6.379939079284668, |
| "learning_rate": 3.900533043523263e-06, |
| "loss": 0.4591, |
| "step": 19200 |
| }, |
| { |
| "epoch": 1.5059969724022824, |
| "grad_norm": 2.010749578475952, |
| "learning_rate": 3.883282444064931e-06, |
| "loss": 0.4877, |
| "step": 19400 |
| }, |
| { |
| "epoch": 1.5215231145441135, |
| "grad_norm": 3.2813642024993896, |
| "learning_rate": 3.8660318446066e-06, |
| "loss": 0.5329, |
| "step": 19600 |
| }, |
| { |
| "epoch": 1.537049256685945, |
| "grad_norm": 2.776346206665039, |
| "learning_rate": 3.848781245148269e-06, |
| "loss": 0.5296, |
| "step": 19800 |
| }, |
| { |
| "epoch": 1.5525753988277762, |
| "grad_norm": 3.623480796813965, |
| "learning_rate": 3.831530645689938e-06, |
| "loss": 0.4708, |
| "step": 20000 |
| }, |
| { |
| "epoch": 1.5525753988277762, |
| "eval_loss": 0.1156494989991188, |
| "eval_runtime": 1531.5832, |
| "eval_samples_per_second": 15.817, |
| "eval_steps_per_second": 1.978, |
| "eval_wer": 0.13344445711797095, |
| "step": 20000 |
| }, |
| { |
| "epoch": 1.5681015409696075, |
| "grad_norm": 1.7083483934402466, |
| "learning_rate": 3.814280046231607e-06, |
| "loss": 0.4518, |
| "step": 20200 |
| }, |
| { |
| "epoch": 1.5836276831114389, |
| "grad_norm": 1.7234033346176147, |
| "learning_rate": 3.7970294467732756e-06, |
| "loss": 0.4809, |
| "step": 20400 |
| }, |
| { |
| "epoch": 1.5991538252532702, |
| "grad_norm": 3.821613311767578, |
| "learning_rate": 3.7797788473149445e-06, |
| "loss": 0.4995, |
| "step": 20600 |
| }, |
| { |
| "epoch": 1.6146799673951016, |
| "grad_norm": 3.65883469581604, |
| "learning_rate": 3.7625282478566134e-06, |
| "loss": 0.5251, |
| "step": 20800 |
| }, |
| { |
| "epoch": 1.6302061095369327, |
| "grad_norm": 2.009864091873169, |
| "learning_rate": 3.7452776483982823e-06, |
| "loss": 0.4664, |
| "step": 21000 |
| }, |
| { |
| "epoch": 1.6457322516787642, |
| "grad_norm": 3.035489559173584, |
| "learning_rate": 3.728027048939951e-06, |
| "loss": 0.431, |
| "step": 21200 |
| }, |
| { |
| "epoch": 1.6612583938205954, |
| "grad_norm": 1.5337060689926147, |
| "learning_rate": 3.7107764494816197e-06, |
| "loss": 0.4236, |
| "step": 21400 |
| }, |
| { |
| "epoch": 1.6767845359624267, |
| "grad_norm": 11.754433631896973, |
| "learning_rate": 3.6935258500232886e-06, |
| "loss": 0.4593, |
| "step": 21600 |
| }, |
| { |
| "epoch": 1.692310678104258, |
| "grad_norm": 4.189357757568359, |
| "learning_rate": 3.6762752505649575e-06, |
| "loss": 0.4892, |
| "step": 21800 |
| }, |
| { |
| "epoch": 1.7078368202460894, |
| "grad_norm": 6.305657863616943, |
| "learning_rate": 3.6590246511066264e-06, |
| "loss": 0.5148, |
| "step": 22000 |
| }, |
| { |
| "epoch": 1.7078368202460894, |
| "eval_loss": 0.11423058062791824, |
| "eval_runtime": 1553.1963, |
| "eval_samples_per_second": 15.597, |
| "eval_steps_per_second": 1.95, |
| "eval_wer": 0.12321260071133047, |
| "step": 22000 |
| }, |
| { |
| "epoch": 1.7233629623879207, |
| "grad_norm": 1.5844590663909912, |
| "learning_rate": 3.6417740516482953e-06, |
| "loss": 0.5071, |
| "step": 22200 |
| }, |
| { |
| "epoch": 1.7388891045297519, |
| "grad_norm": 6.1174139976501465, |
| "learning_rate": 3.6245234521899642e-06, |
| "loss": 0.4507, |
| "step": 22400 |
| }, |
| { |
| "epoch": 1.7544152466715834, |
| "grad_norm": 3.947718381881714, |
| "learning_rate": 3.6072728527316327e-06, |
| "loss": 0.4815, |
| "step": 22600 |
| }, |
| { |
| "epoch": 1.7699413888134146, |
| "grad_norm": 2.0616278648376465, |
| "learning_rate": 3.590108506270593e-06, |
| "loss": 0.4379, |
| "step": 22800 |
| }, |
| { |
| "epoch": 1.785467530955246, |
| "grad_norm": 2.3514456748962402, |
| "learning_rate": 3.572857906812262e-06, |
| "loss": 0.4652, |
| "step": 23000 |
| }, |
| { |
| "epoch": 1.8009936730970773, |
| "grad_norm": 3.154201030731201, |
| "learning_rate": 3.555607307353931e-06, |
| "loss": 0.4038, |
| "step": 23200 |
| }, |
| { |
| "epoch": 1.8165198152389084, |
| "grad_norm": 2.877927303314209, |
| "learning_rate": 3.5383567078955998e-06, |
| "loss": 0.5004, |
| "step": 23400 |
| }, |
| { |
| "epoch": 1.83204595738074, |
| "grad_norm": 3.221912384033203, |
| "learning_rate": 3.5211061084372682e-06, |
| "loss": 0.5206, |
| "step": 23600 |
| }, |
| { |
| "epoch": 1.847572099522571, |
| "grad_norm": 1.4989858865737915, |
| "learning_rate": 3.503855508978937e-06, |
| "loss": 0.5304, |
| "step": 23800 |
| }, |
| { |
| "epoch": 1.8630982416644024, |
| "grad_norm": 1.698549509048462, |
| "learning_rate": 3.486604909520606e-06, |
| "loss": 0.4498, |
| "step": 24000 |
| }, |
| { |
| "epoch": 1.8630982416644024, |
| "eval_loss": 0.11143360286951065, |
| "eval_runtime": 1544.1835, |
| "eval_samples_per_second": 15.688, |
| "eval_steps_per_second": 1.962, |
| "eval_wer": 0.1197829716193656, |
| "step": 24000 |
| }, |
| { |
| "epoch": 1.8786243838062338, |
| "grad_norm": 7.132524013519287, |
| "learning_rate": 3.469354310062275e-06, |
| "loss": 0.4488, |
| "step": 24200 |
| }, |
| { |
| "epoch": 1.894150525948065, |
| "grad_norm": 10.912678718566895, |
| "learning_rate": 3.452103710603944e-06, |
| "loss": 0.4852, |
| "step": 24400 |
| }, |
| { |
| "epoch": 1.9096766680898964, |
| "grad_norm": 2.820404529571533, |
| "learning_rate": 3.4348531111456124e-06, |
| "loss": 0.499, |
| "step": 24600 |
| }, |
| { |
| "epoch": 1.9252028102317276, |
| "grad_norm": 1.029526948928833, |
| "learning_rate": 3.417688764684573e-06, |
| "loss": 0.4281, |
| "step": 24800 |
| }, |
| { |
| "epoch": 1.9407289523735591, |
| "grad_norm": 9.125391006469727, |
| "learning_rate": 3.400438165226242e-06, |
| "loss": 0.4625, |
| "step": 25000 |
| }, |
| { |
| "epoch": 1.9562550945153903, |
| "grad_norm": 1.7483952045440674, |
| "learning_rate": 3.383360071762494e-06, |
| "loss": 0.3859, |
| "step": 25200 |
| }, |
| { |
| "epoch": 1.9717812366572216, |
| "grad_norm": 1.5715336799621582, |
| "learning_rate": 3.3661094723041627e-06, |
| "loss": 0.4456, |
| "step": 25400 |
| }, |
| { |
| "epoch": 1.987307378799053, |
| "grad_norm": 2.644300699234009, |
| "learning_rate": 3.3488588728458316e-06, |
| "loss": 0.4182, |
| "step": 25600 |
| }, |
| { |
| "epoch": 2.0027947055855297, |
| "grad_norm": 3.5517842769622803, |
| "learning_rate": 3.3316082733875005e-06, |
| "loss": 0.502, |
| "step": 25800 |
| }, |
| { |
| "epoch": 2.018320847727361, |
| "grad_norm": 3.494882583618164, |
| "learning_rate": 3.3143576739291694e-06, |
| "loss": 0.4499, |
| "step": 26000 |
| }, |
| { |
| "epoch": 2.018320847727361, |
| "eval_loss": 0.11194167286157608, |
| "eval_runtime": 1499.1763, |
| "eval_samples_per_second": 16.159, |
| "eval_steps_per_second": 2.02, |
| "eval_wer": 0.12530718900030072, |
| "step": 26000 |
| }, |
| { |
| "epoch": 2.0338469898691924, |
| "grad_norm": 3.632197618484497, |
| "learning_rate": 3.2971070744708383e-06, |
| "loss": 0.4946, |
| "step": 26200 |
| }, |
| { |
| "epoch": 2.0493731320110236, |
| "grad_norm": 4.505156517028809, |
| "learning_rate": 3.2798564750125068e-06, |
| "loss": 0.4587, |
| "step": 26400 |
| }, |
| { |
| "epoch": 2.0648992741528547, |
| "grad_norm": 6.850549221038818, |
| "learning_rate": 3.2626058755541757e-06, |
| "loss": 0.4937, |
| "step": 26600 |
| }, |
| { |
| "epoch": 2.0804254162946862, |
| "grad_norm": 1.3784022331237793, |
| "learning_rate": 3.2453552760958446e-06, |
| "loss": 0.4512, |
| "step": 26800 |
| }, |
| { |
| "epoch": 2.0959515584365174, |
| "grad_norm": 3.1638882160186768, |
| "learning_rate": 3.2281046766375135e-06, |
| "loss": 0.3987, |
| "step": 27000 |
| }, |
| { |
| "epoch": 2.111477700578349, |
| "grad_norm": 3.5049760341644287, |
| "learning_rate": 3.2108540771791824e-06, |
| "loss": 0.4348, |
| "step": 27200 |
| }, |
| { |
| "epoch": 2.12700384272018, |
| "grad_norm": 1.3930319547653198, |
| "learning_rate": 3.1936034777208513e-06, |
| "loss": 0.4324, |
| "step": 27400 |
| }, |
| { |
| "epoch": 2.142529984862011, |
| "grad_norm": 3.2007176876068115, |
| "learning_rate": 3.17635287826252e-06, |
| "loss": 0.4076, |
| "step": 27600 |
| }, |
| { |
| "epoch": 2.1580561270038428, |
| "grad_norm": 1.4413760900497437, |
| "learning_rate": 3.1591022788041887e-06, |
| "loss": 0.364, |
| "step": 27800 |
| }, |
| { |
| "epoch": 2.173582269145674, |
| "grad_norm": 1.6212908029556274, |
| "learning_rate": 3.1418516793458576e-06, |
| "loss": 0.4502, |
| "step": 28000 |
| }, |
| { |
| "epoch": 2.173582269145674, |
| "eval_loss": 0.11298074573278427, |
| "eval_runtime": 1460.6953, |
| "eval_samples_per_second": 16.585, |
| "eval_steps_per_second": 2.074, |
| "eval_wer": 0.12457097232447453, |
| "step": 28000 |
| }, |
| { |
| "epoch": 2.1891084112875054, |
| "grad_norm": 2.5020694732666016, |
| "learning_rate": 3.1246010798875265e-06, |
| "loss": 0.4184, |
| "step": 28200 |
| }, |
| { |
| "epoch": 2.2046345534293366, |
| "grad_norm": 6.708138942718506, |
| "learning_rate": 3.1073504804291954e-06, |
| "loss": 0.2974, |
| "step": 28400 |
| }, |
| { |
| "epoch": 2.220160695571168, |
| "grad_norm": 5.524842739105225, |
| "learning_rate": 3.090099880970864e-06, |
| "loss": 0.4081, |
| "step": 28600 |
| }, |
| { |
| "epoch": 2.2356868377129993, |
| "grad_norm": 6.652355670928955, |
| "learning_rate": 3.072849281512533e-06, |
| "loss": 0.3646, |
| "step": 28800 |
| }, |
| { |
| "epoch": 2.251212979854831, |
| "grad_norm": 3.2332844734191895, |
| "learning_rate": 3.0555986820542017e-06, |
| "loss": 0.3628, |
| "step": 29000 |
| }, |
| { |
| "epoch": 2.266739121996662, |
| "grad_norm": 2.7129597663879395, |
| "learning_rate": 3.0383480825958707e-06, |
| "loss": 0.3407, |
| "step": 29200 |
| }, |
| { |
| "epoch": 2.282265264138493, |
| "grad_norm": 3.1002357006073, |
| "learning_rate": 3.0210974831375396e-06, |
| "loss": 0.3257, |
| "step": 29400 |
| }, |
| { |
| "epoch": 2.2977914062803246, |
| "grad_norm": 1.7787259817123413, |
| "learning_rate": 3.003846883679208e-06, |
| "loss": 0.3418, |
| "step": 29600 |
| }, |
| { |
| "epoch": 2.3133175484221558, |
| "grad_norm": 2.824995517730713, |
| "learning_rate": 2.986596284220877e-06, |
| "loss": 0.3289, |
| "step": 29800 |
| }, |
| { |
| "epoch": 2.3288436905639873, |
| "grad_norm": 4.119309425354004, |
| "learning_rate": 2.969345684762546e-06, |
| "loss": 0.3485, |
| "step": 30000 |
| }, |
| { |
| "epoch": 2.3288436905639873, |
| "eval_loss": 0.1113351359963417, |
| "eval_runtime": 1572.7013, |
| "eval_samples_per_second": 15.403, |
| "eval_steps_per_second": 1.926, |
| "eval_wer": 0.11538641006231919, |
| "step": 30000 |
| }, |
| { |
| "epoch": 2.3443698327058184, |
| "grad_norm": 3.298201322555542, |
| "learning_rate": 2.9520950853042148e-06, |
| "loss": 0.3738, |
| "step": 30200 |
| }, |
| { |
| "epoch": 2.3598959748476496, |
| "grad_norm": 2.8228886127471924, |
| "learning_rate": 2.934930738843175e-06, |
| "loss": 0.3146, |
| "step": 30400 |
| }, |
| { |
| "epoch": 2.375422116989481, |
| "grad_norm": 1.6474792957305908, |
| "learning_rate": 2.9176801393848436e-06, |
| "loss": 0.3198, |
| "step": 30600 |
| }, |
| { |
| "epoch": 2.3909482591313123, |
| "grad_norm": 2.8176417350769043, |
| "learning_rate": 2.9004295399265125e-06, |
| "loss": 0.3763, |
| "step": 30800 |
| }, |
| { |
| "epoch": 2.406474401273144, |
| "grad_norm": 10.521832466125488, |
| "learning_rate": 2.8831789404681814e-06, |
| "loss": 0.341, |
| "step": 31000 |
| }, |
| { |
| "epoch": 2.422000543414975, |
| "grad_norm": 2.951089859008789, |
| "learning_rate": 2.8659283410098503e-06, |
| "loss": 0.4033, |
| "step": 31200 |
| }, |
| { |
| "epoch": 2.437526685556806, |
| "grad_norm": 1.2934590578079224, |
| "learning_rate": 2.8486777415515192e-06, |
| "loss": 0.319, |
| "step": 31400 |
| }, |
| { |
| "epoch": 2.4530528276986376, |
| "grad_norm": 1.497093915939331, |
| "learning_rate": 2.8314271420931877e-06, |
| "loss": 0.3251, |
| "step": 31600 |
| }, |
| { |
| "epoch": 2.4685789698404688, |
| "grad_norm": 1.2919329404830933, |
| "learning_rate": 2.8141765426348566e-06, |
| "loss": 0.314, |
| "step": 31800 |
| }, |
| { |
| "epoch": 2.4841051119823003, |
| "grad_norm": 4.007678985595703, |
| "learning_rate": 2.7969259431765255e-06, |
| "loss": 0.3248, |
| "step": 32000 |
| }, |
| { |
| "epoch": 2.4841051119823003, |
| "eval_loss": 0.10955296456813812, |
| "eval_runtime": 1592.1004, |
| "eval_samples_per_second": 15.216, |
| "eval_steps_per_second": 1.903, |
| "eval_wer": 0.11213824282707203, |
| "step": 32000 |
| }, |
| { |
| "epoch": 2.4996312541241315, |
| "grad_norm": 2.58063006401062, |
| "learning_rate": 2.7796753437181944e-06, |
| "loss": 0.3809, |
| "step": 32200 |
| }, |
| { |
| "epoch": 2.5151573962659626, |
| "grad_norm": 4.455554485321045, |
| "learning_rate": 2.7624247442598633e-06, |
| "loss": 0.3105, |
| "step": 32400 |
| }, |
| { |
| "epoch": 2.530683538407794, |
| "grad_norm": 1.4546878337860107, |
| "learning_rate": 2.7452603977988237e-06, |
| "loss": 0.3197, |
| "step": 32600 |
| }, |
| { |
| "epoch": 2.5462096805496257, |
| "grad_norm": 1.7924015522003174, |
| "learning_rate": 2.7280097983404926e-06, |
| "loss": 0.3739, |
| "step": 32800 |
| }, |
| { |
| "epoch": 2.561735822691457, |
| "grad_norm": 1.9410858154296875, |
| "learning_rate": 2.7107591988821615e-06, |
| "loss": 0.3624, |
| "step": 33000 |
| }, |
| { |
| "epoch": 2.577261964833288, |
| "grad_norm": 4.357357025146484, |
| "learning_rate": 2.6935085994238304e-06, |
| "loss": 0.353, |
| "step": 33200 |
| }, |
| { |
| "epoch": 2.5927881069751195, |
| "grad_norm": 2.347006320953369, |
| "learning_rate": 2.6762579999654993e-06, |
| "loss": 0.2845, |
| "step": 33400 |
| }, |
| { |
| "epoch": 2.6083142491169506, |
| "grad_norm": 3.3668770790100098, |
| "learning_rate": 2.659007400507168e-06, |
| "loss": 0.3098, |
| "step": 33600 |
| }, |
| { |
| "epoch": 2.623840391258782, |
| "grad_norm": 8.008859634399414, |
| "learning_rate": 2.6417568010488363e-06, |
| "loss": 0.3462, |
| "step": 33800 |
| }, |
| { |
| "epoch": 2.6393665334006133, |
| "grad_norm": 2.9489552974700928, |
| "learning_rate": 2.6245062015905056e-06, |
| "loss": 0.3597, |
| "step": 34000 |
| }, |
| { |
| "epoch": 2.6393665334006133, |
| "eval_loss": 0.1078966036438942, |
| "eval_runtime": 1558.3985, |
| "eval_samples_per_second": 15.545, |
| "eval_steps_per_second": 1.944, |
| "eval_wer": 0.11223934300438619, |
| "step": 34000 |
| }, |
| { |
| "epoch": 2.6548926755424445, |
| "grad_norm": 3.753852128982544, |
| "learning_rate": 2.607341855129466e-06, |
| "loss": 0.3436, |
| "step": 34200 |
| }, |
| { |
| "epoch": 2.670418817684276, |
| "grad_norm": 1.924181342124939, |
| "learning_rate": 2.590091255671135e-06, |
| "loss": 0.2933, |
| "step": 34400 |
| }, |
| { |
| "epoch": 2.685944959826107, |
| "grad_norm": 2.9683339595794678, |
| "learning_rate": 2.5728406562128037e-06, |
| "loss": 0.327, |
| "step": 34600 |
| }, |
| { |
| "epoch": 2.7014711019679387, |
| "grad_norm": 3.8935837745666504, |
| "learning_rate": 2.5555900567544722e-06, |
| "loss": 0.2978, |
| "step": 34800 |
| }, |
| { |
| "epoch": 2.71699724410977, |
| "grad_norm": 1.7941005229949951, |
| "learning_rate": 2.538339457296141e-06, |
| "loss": 0.3582, |
| "step": 35000 |
| }, |
| { |
| "epoch": 2.732523386251601, |
| "grad_norm": 3.8114750385284424, |
| "learning_rate": 2.52108885783781e-06, |
| "loss": 0.3341, |
| "step": 35200 |
| }, |
| { |
| "epoch": 2.7480495283934325, |
| "grad_norm": 4.2053446769714355, |
| "learning_rate": 2.503838258379479e-06, |
| "loss": 0.2964, |
| "step": 35400 |
| }, |
| { |
| "epoch": 2.7635756705352637, |
| "grad_norm": 1.7080585956573486, |
| "learning_rate": 2.486587658921148e-06, |
| "loss": 0.2737, |
| "step": 35600 |
| }, |
| { |
| "epoch": 2.779101812677095, |
| "grad_norm": 2.3851122856140137, |
| "learning_rate": 2.4693370594628168e-06, |
| "loss": 0.2916, |
| "step": 35800 |
| }, |
| { |
| "epoch": 2.7946279548189263, |
| "grad_norm": 1.548720121383667, |
| "learning_rate": 2.4520864600044853e-06, |
| "loss": 0.2844, |
| "step": 36000 |
| }, |
| { |
| "epoch": 2.7946279548189263, |
| "eval_loss": 0.10499973595142365, |
| "eval_runtime": 1489.5099, |
| "eval_samples_per_second": 16.264, |
| "eval_steps_per_second": 2.034, |
| "eval_wer": 0.11226785843901327, |
| "step": 36000 |
| }, |
| { |
| "epoch": 2.8101540969607575, |
| "grad_norm": 3.8322744369506836, |
| "learning_rate": 2.434835860546154e-06, |
| "loss": 0.3067, |
| "step": 36200 |
| }, |
| { |
| "epoch": 2.825680239102589, |
| "grad_norm": 0.8839676380157471, |
| "learning_rate": 2.417585261087823e-06, |
| "loss": 0.2895, |
| "step": 36400 |
| }, |
| { |
| "epoch": 2.84120638124442, |
| "grad_norm": 1.9399739503860474, |
| "learning_rate": 2.4003346616294916e-06, |
| "loss": 0.3003, |
| "step": 36600 |
| }, |
| { |
| "epoch": 2.8567325233862517, |
| "grad_norm": 2.682796001434326, |
| "learning_rate": 2.3830840621711605e-06, |
| "loss": 0.2886, |
| "step": 36800 |
| }, |
| { |
| "epoch": 2.872258665528083, |
| "grad_norm": 1.7100605964660645, |
| "learning_rate": 2.36583346271283e-06, |
| "loss": 0.2891, |
| "step": 37000 |
| }, |
| { |
| "epoch": 2.887784807669914, |
| "grad_norm": 2.3585317134857178, |
| "learning_rate": 2.3485828632544983e-06, |
| "loss": 0.3101, |
| "step": 37200 |
| }, |
| { |
| "epoch": 2.9033109498117455, |
| "grad_norm": 1.7011852264404297, |
| "learning_rate": 2.331332263796167e-06, |
| "loss": 0.3289, |
| "step": 37400 |
| }, |
| { |
| "epoch": 2.918837091953577, |
| "grad_norm": 5.483014106750488, |
| "learning_rate": 2.3140816643378357e-06, |
| "loss": 0.2616, |
| "step": 37600 |
| }, |
| { |
| "epoch": 2.9343632340954082, |
| "grad_norm": 34.07948303222656, |
| "learning_rate": 2.2968310648795046e-06, |
| "loss": 0.3142, |
| "step": 37800 |
| }, |
| { |
| "epoch": 2.9498893762372393, |
| "grad_norm": 1.9521312713623047, |
| "learning_rate": 2.2795804654211735e-06, |
| "loss": 0.3192, |
| "step": 38000 |
| }, |
| { |
| "epoch": 2.9498893762372393, |
| "eval_loss": 0.10090235620737076, |
| "eval_runtime": 1467.3564, |
| "eval_samples_per_second": 16.509, |
| "eval_steps_per_second": 2.064, |
| "eval_wer": 0.11597745725277119, |
| "step": 38000 |
| }, |
| { |
| "epoch": 2.965415518379071, |
| "grad_norm": 3.245943546295166, |
| "learning_rate": 2.2623298659628424e-06, |
| "loss": 0.278, |
| "step": 38200 |
| }, |
| { |
| "epoch": 2.980941660520902, |
| "grad_norm": 2.6015217304229736, |
| "learning_rate": 2.2450792665045113e-06, |
| "loss": 0.3133, |
| "step": 38400 |
| }, |
| { |
| "epoch": 2.9964678026627336, |
| "grad_norm": 0.971225917339325, |
| "learning_rate": 2.2278286670461802e-06, |
| "loss": 0.3025, |
| "step": 38600 |
| }, |
| { |
| "epoch": 3.01195512944921, |
| "grad_norm": 1.830532193183899, |
| "learning_rate": 2.2106643205851405e-06, |
| "loss": 0.3117, |
| "step": 38800 |
| }, |
| { |
| "epoch": 3.0274812715910415, |
| "grad_norm": 3.5241925716400146, |
| "learning_rate": 2.1934137211268095e-06, |
| "loss": 0.2681, |
| "step": 39000 |
| }, |
| { |
| "epoch": 3.0430074137328726, |
| "grad_norm": 0.8706326484680176, |
| "learning_rate": 2.176163121668478e-06, |
| "loss": 0.2982, |
| "step": 39200 |
| }, |
| { |
| "epoch": 3.058533555874704, |
| "grad_norm": 4.333695411682129, |
| "learning_rate": 2.158912522210147e-06, |
| "loss": 0.3, |
| "step": 39400 |
| }, |
| { |
| "epoch": 3.0740596980165353, |
| "grad_norm": 3.2154648303985596, |
| "learning_rate": 2.1416619227518158e-06, |
| "loss": 0.2708, |
| "step": 39600 |
| }, |
| { |
| "epoch": 3.0895858401583665, |
| "grad_norm": 4.098783493041992, |
| "learning_rate": 2.1244113232934847e-06, |
| "loss": 0.2772, |
| "step": 39800 |
| }, |
| { |
| "epoch": 3.105111982300198, |
| "grad_norm": 2.571160078048706, |
| "learning_rate": 2.1071607238351536e-06, |
| "loss": 0.2805, |
| "step": 40000 |
| }, |
| { |
| "epoch": 3.105111982300198, |
| "eval_loss": 0.10206975787878036, |
| "eval_runtime": 1458.9109, |
| "eval_samples_per_second": 16.605, |
| "eval_steps_per_second": 2.076, |
| "eval_wer": 0.11133462603303643, |
| "step": 40000 |
| }, |
| { |
| "epoch": 3.120638124442029, |
| "grad_norm": 2.1225316524505615, |
| "learning_rate": 2.089910124376822e-06, |
| "loss": 0.2797, |
| "step": 40200 |
| }, |
| { |
| "epoch": 3.1361642665838607, |
| "grad_norm": 1.6316463947296143, |
| "learning_rate": 2.072659524918491e-06, |
| "loss": 0.333, |
| "step": 40400 |
| }, |
| { |
| "epoch": 3.151690408725692, |
| "grad_norm": 5.42690372467041, |
| "learning_rate": 2.05540892546016e-06, |
| "loss": 0.2893, |
| "step": 40600 |
| }, |
| { |
| "epoch": 3.167216550867523, |
| "grad_norm": 1.4130024909973145, |
| "learning_rate": 2.038158326001829e-06, |
| "loss": 0.305, |
| "step": 40800 |
| }, |
| { |
| "epoch": 3.1827426930093545, |
| "grad_norm": 2.8819572925567627, |
| "learning_rate": 2.0209077265434977e-06, |
| "loss": 0.3095, |
| "step": 41000 |
| }, |
| { |
| "epoch": 3.1982688351511857, |
| "grad_norm": 1.9537643194198608, |
| "learning_rate": 2.003657127085166e-06, |
| "loss": 0.2849, |
| "step": 41200 |
| }, |
| { |
| "epoch": 3.213794977293017, |
| "grad_norm": 2.2093582153320312, |
| "learning_rate": 1.986406527626835e-06, |
| "loss": 0.2819, |
| "step": 41400 |
| }, |
| { |
| "epoch": 3.2293211194348483, |
| "grad_norm": 3.2234604358673096, |
| "learning_rate": 1.969155928168504e-06, |
| "loss": 0.2599, |
| "step": 41600 |
| }, |
| { |
| "epoch": 3.24484726157668, |
| "grad_norm": 1.1582082509994507, |
| "learning_rate": 1.951905328710173e-06, |
| "loss": 0.2958, |
| "step": 41800 |
| }, |
| { |
| "epoch": 3.260373403718511, |
| "grad_norm": 1.816443681716919, |
| "learning_rate": 1.934654729251842e-06, |
| "loss": 0.2699, |
| "step": 42000 |
| }, |
| { |
| "epoch": 3.260373403718511, |
| "eval_loss": 0.1001201719045639, |
| "eval_runtime": 1453.9545, |
| "eval_samples_per_second": 16.661, |
| "eval_steps_per_second": 2.083, |
| "eval_wer": 0.11127241053930464, |
| "step": 42000 |
| }, |
| { |
| "epoch": 3.2758995458603426, |
| "grad_norm": 1.9462873935699463, |
| "learning_rate": 1.9174041297935107e-06, |
| "loss": 0.2846, |
| "step": 42200 |
| }, |
| { |
| "epoch": 3.2914256880021737, |
| "grad_norm": 5.209357261657715, |
| "learning_rate": 1.9001535303351792e-06, |
| "loss": 0.3215, |
| "step": 42400 |
| }, |
| { |
| "epoch": 3.306951830144005, |
| "grad_norm": 3.4322330951690674, |
| "learning_rate": 1.8829029308768481e-06, |
| "loss": 0.272, |
| "step": 42600 |
| }, |
| { |
| "epoch": 3.3224779722858364, |
| "grad_norm": 1.4792200326919556, |
| "learning_rate": 1.8657385844158084e-06, |
| "loss": 0.2739, |
| "step": 42800 |
| }, |
| { |
| "epoch": 3.3380041144276675, |
| "grad_norm": 1.1712541580200195, |
| "learning_rate": 1.8484879849574774e-06, |
| "loss": 0.2469, |
| "step": 43000 |
| }, |
| { |
| "epoch": 3.353530256569499, |
| "grad_norm": 5.4289422035217285, |
| "learning_rate": 1.8312373854991463e-06, |
| "loss": 0.264, |
| "step": 43200 |
| }, |
| { |
| "epoch": 3.3690563987113302, |
| "grad_norm": 1.5609965324401855, |
| "learning_rate": 1.813986786040815e-06, |
| "loss": 0.2814, |
| "step": 43400 |
| }, |
| { |
| "epoch": 3.3845825408531613, |
| "grad_norm": 2.766329526901245, |
| "learning_rate": 1.7967361865824839e-06, |
| "loss": 0.2827, |
| "step": 43600 |
| }, |
| { |
| "epoch": 3.400108682994993, |
| "grad_norm": 2.752129077911377, |
| "learning_rate": 1.7794855871241526e-06, |
| "loss": 0.2565, |
| "step": 43800 |
| }, |
| { |
| "epoch": 3.415634825136824, |
| "grad_norm": 5.707192420959473, |
| "learning_rate": 1.7622349876658215e-06, |
| "loss": 0.3088, |
| "step": 44000 |
| }, |
| { |
| "epoch": 3.415634825136824, |
| "eval_loss": 0.09880851954221725, |
| "eval_runtime": 1490.9962, |
| "eval_samples_per_second": 16.248, |
| "eval_steps_per_second": 2.032, |
| "eval_wer": 0.11230674312259563, |
| "step": 44000 |
| }, |
| { |
| "epoch": 3.4311609672786556, |
| "grad_norm": 4.677556991577148, |
| "learning_rate": 1.7450706412047822e-06, |
| "loss": 0.2729, |
| "step": 44200 |
| }, |
| { |
| "epoch": 3.4466871094204867, |
| "grad_norm": 2.3275015354156494, |
| "learning_rate": 1.7278200417464507e-06, |
| "loss": 0.2528, |
| "step": 44400 |
| }, |
| { |
| "epoch": 3.462213251562318, |
| "grad_norm": 2.021616220474243, |
| "learning_rate": 1.7105694422881196e-06, |
| "loss": 0.2281, |
| "step": 44600 |
| }, |
| { |
| "epoch": 3.4777393937041494, |
| "grad_norm": 4.585258960723877, |
| "learning_rate": 1.6933188428297883e-06, |
| "loss": 0.2812, |
| "step": 44800 |
| }, |
| { |
| "epoch": 3.4932655358459805, |
| "grad_norm": 8.669875144958496, |
| "learning_rate": 1.6760682433714572e-06, |
| "loss": 0.2648, |
| "step": 45000 |
| }, |
| { |
| "epoch": 3.508791677987812, |
| "grad_norm": 4.819867134094238, |
| "learning_rate": 1.6588176439131261e-06, |
| "loss": 0.26, |
| "step": 45200 |
| }, |
| { |
| "epoch": 3.5243178201296432, |
| "grad_norm": 3.6191158294677734, |
| "learning_rate": 1.6415670444547948e-06, |
| "loss": 0.2834, |
| "step": 45400 |
| }, |
| { |
| "epoch": 3.5398439622714744, |
| "grad_norm": 1.5337854623794556, |
| "learning_rate": 1.6243164449964637e-06, |
| "loss": 0.2768, |
| "step": 45600 |
| }, |
| { |
| "epoch": 3.555370104413306, |
| "grad_norm": 3.030637741088867, |
| "learning_rate": 1.6070658455381327e-06, |
| "loss": 0.2528, |
| "step": 45800 |
| }, |
| { |
| "epoch": 3.5708962465551375, |
| "grad_norm": 0.9973781108856201, |
| "learning_rate": 1.5898152460798013e-06, |
| "loss": 0.2805, |
| "step": 46000 |
| }, |
| { |
| "epoch": 3.5708962465551375, |
| "eval_loss": 0.09769914299249649, |
| "eval_runtime": 1578.5646, |
| "eval_samples_per_second": 15.346, |
| "eval_steps_per_second": 1.919, |
| "eval_wer": 0.11198011178050374, |
| "step": 46000 |
| }, |
| { |
| "epoch": 3.5864223886969686, |
| "grad_norm": 4.177305698394775, |
| "learning_rate": 1.5725646466214703e-06, |
| "loss": 0.2928, |
| "step": 46200 |
| }, |
| { |
| "epoch": 3.6019485308387997, |
| "grad_norm": 4.166442394256592, |
| "learning_rate": 1.555314047163139e-06, |
| "loss": 0.2421, |
| "step": 46400 |
| }, |
| { |
| "epoch": 3.6174746729806313, |
| "grad_norm": 2.1789302825927734, |
| "learning_rate": 1.5380634477048079e-06, |
| "loss": 0.294, |
| "step": 46600 |
| }, |
| { |
| "epoch": 3.6330008151224624, |
| "grad_norm": 4.862135887145996, |
| "learning_rate": 1.5208128482464768e-06, |
| "loss": 0.2575, |
| "step": 46800 |
| }, |
| { |
| "epoch": 3.648526957264294, |
| "grad_norm": 2.7330503463745117, |
| "learning_rate": 1.5035622487881455e-06, |
| "loss": 0.2364, |
| "step": 47000 |
| }, |
| { |
| "epoch": 3.664053099406125, |
| "grad_norm": 1.8307307958602905, |
| "learning_rate": 1.4863116493298144e-06, |
| "loss": 0.2581, |
| "step": 47200 |
| }, |
| { |
| "epoch": 3.6795792415479562, |
| "grad_norm": 4.16917610168457, |
| "learning_rate": 1.4690610498714833e-06, |
| "loss": 0.2418, |
| "step": 47400 |
| }, |
| { |
| "epoch": 3.695105383689788, |
| "grad_norm": 4.166003704071045, |
| "learning_rate": 1.451810450413152e-06, |
| "loss": 0.2554, |
| "step": 47600 |
| }, |
| { |
| "epoch": 3.710631525831619, |
| "grad_norm": 6.75455904006958, |
| "learning_rate": 1.434559850954821e-06, |
| "loss": 0.2621, |
| "step": 47800 |
| }, |
| { |
| "epoch": 3.7261576679734505, |
| "grad_norm": 3.2744832038879395, |
| "learning_rate": 1.4173092514964896e-06, |
| "loss": 0.2394, |
| "step": 48000 |
| }, |
| { |
| "epoch": 3.7261576679734505, |
| "eval_loss": 0.09708496928215027, |
| "eval_runtime": 1602.7102, |
| "eval_samples_per_second": 15.115, |
| "eval_steps_per_second": 1.89, |
| "eval_wer": 0.1107928327751221, |
| "step": 48000 |
| }, |
| { |
| "epoch": 3.7416838101152816, |
| "grad_norm": 1.581204891204834, |
| "learning_rate": 1.4000586520381585e-06, |
| "loss": 0.2656, |
| "step": 48200 |
| }, |
| { |
| "epoch": 3.7572099522571127, |
| "grad_norm": 0.8444304466247559, |
| "learning_rate": 1.3828943055771188e-06, |
| "loss": 0.2759, |
| "step": 48400 |
| }, |
| { |
| "epoch": 3.7727360943989443, |
| "grad_norm": 3.9065730571746826, |
| "learning_rate": 1.3656437061187877e-06, |
| "loss": 0.2588, |
| "step": 48600 |
| }, |
| { |
| "epoch": 3.7882622365407754, |
| "grad_norm": 4.403543472290039, |
| "learning_rate": 1.3483931066604566e-06, |
| "loss": 0.2702, |
| "step": 48800 |
| }, |
| { |
| "epoch": 3.803788378682607, |
| "grad_norm": 0.8027508854866028, |
| "learning_rate": 1.3311425072021253e-06, |
| "loss": 0.2223, |
| "step": 49000 |
| }, |
| { |
| "epoch": 3.819314520824438, |
| "grad_norm": 8.521760940551758, |
| "learning_rate": 1.3138919077437942e-06, |
| "loss": 0.2651, |
| "step": 49200 |
| }, |
| { |
| "epoch": 3.8348406629662692, |
| "grad_norm": 2.732095241546631, |
| "learning_rate": 1.2966413082854632e-06, |
| "loss": 0.2482, |
| "step": 49400 |
| }, |
| { |
| "epoch": 3.850366805108101, |
| "grad_norm": 3.1505069732666016, |
| "learning_rate": 1.2793907088271319e-06, |
| "loss": 0.2212, |
| "step": 49600 |
| }, |
| { |
| "epoch": 3.865892947249932, |
| "grad_norm": 10.347441673278809, |
| "learning_rate": 1.2621401093688008e-06, |
| "loss": 0.266, |
| "step": 49800 |
| }, |
| { |
| "epoch": 3.8814190893917635, |
| "grad_norm": 3.445387125015259, |
| "learning_rate": 1.2448895099104695e-06, |
| "loss": 0.2486, |
| "step": 50000 |
| }, |
| { |
| "epoch": 3.8814190893917635, |
| "eval_loss": 0.09692510962486267, |
| "eval_runtime": 1572.1321, |
| "eval_samples_per_second": 15.409, |
| "eval_steps_per_second": 1.927, |
| "eval_wer": 0.10840790551540352, |
| "step": 50000 |
| }, |
| { |
| "epoch": 3.8969452315335946, |
| "grad_norm": 1.5172524452209473, |
| "learning_rate": 1.2276389104521384e-06, |
| "loss": 0.2291, |
| "step": 50200 |
| }, |
| { |
| "epoch": 3.9124713736754257, |
| "grad_norm": 12.397218704223633, |
| "learning_rate": 1.210388310993807e-06, |
| "loss": 0.3028, |
| "step": 50400 |
| }, |
| { |
| "epoch": 3.9279975158172573, |
| "grad_norm": 1.5778107643127441, |
| "learning_rate": 1.1932239645327676e-06, |
| "loss": 0.3061, |
| "step": 50600 |
| }, |
| { |
| "epoch": 3.943523657959089, |
| "grad_norm": 1.6338222026824951, |
| "learning_rate": 1.1759733650744365e-06, |
| "loss": 0.2447, |
| "step": 50800 |
| }, |
| { |
| "epoch": 3.95904980010092, |
| "grad_norm": 3.10078763961792, |
| "learning_rate": 1.1587227656161052e-06, |
| "loss": 0.2678, |
| "step": 51000 |
| }, |
| { |
| "epoch": 3.974575942242751, |
| "grad_norm": 4.663175106048584, |
| "learning_rate": 1.1414721661577741e-06, |
| "loss": 0.2581, |
| "step": 51200 |
| }, |
| { |
| "epoch": 3.9901020843845827, |
| "grad_norm": 8.915782928466797, |
| "learning_rate": 1.1242215666994428e-06, |
| "loss": 0.262, |
| "step": 51400 |
| }, |
| { |
| "epoch": 4.0055894111710595, |
| "grad_norm": 2.30519962310791, |
| "learning_rate": 1.1069709672411117e-06, |
| "loss": 0.2746, |
| "step": 51600 |
| }, |
| { |
| "epoch": 4.021115553312891, |
| "grad_norm": 2.362076997756958, |
| "learning_rate": 1.0897203677827806e-06, |
| "loss": 0.2542, |
| "step": 51800 |
| }, |
| { |
| "epoch": 4.036641695454722, |
| "grad_norm": 2.8496739864349365, |
| "learning_rate": 1.0724697683244493e-06, |
| "loss": 0.2577, |
| "step": 52000 |
| }, |
| { |
| "epoch": 4.036641695454722, |
| "eval_loss": 0.09788235276937485, |
| "eval_runtime": 1600.6017, |
| "eval_samples_per_second": 15.135, |
| "eval_steps_per_second": 1.892, |
| "eval_wer": 0.10990366967720527, |
| "step": 52000 |
| }, |
| { |
| "epoch": 4.052167837596553, |
| "grad_norm": 1.7787532806396484, |
| "learning_rate": 1.0552191688661182e-06, |
| "loss": 0.2586, |
| "step": 52200 |
| }, |
| { |
| "epoch": 4.067693979738385, |
| "grad_norm": 7.863888740539551, |
| "learning_rate": 1.0379685694077871e-06, |
| "loss": 0.2433, |
| "step": 52400 |
| }, |
| { |
| "epoch": 4.083220121880216, |
| "grad_norm": 2.5998260974884033, |
| "learning_rate": 1.0207179699494558e-06, |
| "loss": 0.288, |
| "step": 52600 |
| }, |
| { |
| "epoch": 4.098746264022047, |
| "grad_norm": 5.83883810043335, |
| "learning_rate": 1.0035536234884164e-06, |
| "loss": 0.2605, |
| "step": 52800 |
| }, |
| { |
| "epoch": 4.114272406163878, |
| "grad_norm": 1.4180282354354858, |
| "learning_rate": 9.86389277027377e-07, |
| "loss": 0.2759, |
| "step": 53000 |
| }, |
| { |
| "epoch": 4.129798548305709, |
| "grad_norm": 2.4675915241241455, |
| "learning_rate": 9.691386775690456e-07, |
| "loss": 0.2568, |
| "step": 53200 |
| }, |
| { |
| "epoch": 4.145324690447541, |
| "grad_norm": 1.6407579183578491, |
| "learning_rate": 9.518880781107144e-07, |
| "loss": 0.2407, |
| "step": 53400 |
| }, |
| { |
| "epoch": 4.1608508325893725, |
| "grad_norm": 1.6074066162109375, |
| "learning_rate": 9.346374786523832e-07, |
| "loss": 0.2669, |
| "step": 53600 |
| }, |
| { |
| "epoch": 4.176376974731204, |
| "grad_norm": 3.188389301300049, |
| "learning_rate": 9.173868791940521e-07, |
| "loss": 0.2854, |
| "step": 53800 |
| }, |
| { |
| "epoch": 4.191903116873035, |
| "grad_norm": 7.821218013763428, |
| "learning_rate": 9.001362797357209e-07, |
| "loss": 0.2409, |
| "step": 54000 |
| }, |
| { |
| "epoch": 4.191903116873035, |
| "eval_loss": 0.09744942933320999, |
| "eval_runtime": 1614.7753, |
| "eval_samples_per_second": 15.002, |
| "eval_steps_per_second": 1.876, |
| "eval_wer": 0.10502753035597631, |
| "step": 54000 |
| }, |
| { |
| "epoch": 4.207429259014866, |
| "grad_norm": 3.503199815750122, |
| "learning_rate": 8.828856802773897e-07, |
| "loss": 0.2485, |
| "step": 54200 |
| }, |
| { |
| "epoch": 4.222955401156698, |
| "grad_norm": 6.2170233726501465, |
| "learning_rate": 8.656350808190585e-07, |
| "loss": 0.2758, |
| "step": 54400 |
| }, |
| { |
| "epoch": 4.238481543298529, |
| "grad_norm": 2.058504819869995, |
| "learning_rate": 8.483844813607274e-07, |
| "loss": 0.2713, |
| "step": 54600 |
| }, |
| { |
| "epoch": 4.25400768544036, |
| "grad_norm": 2.693246603012085, |
| "learning_rate": 8.311338819023962e-07, |
| "loss": 0.2309, |
| "step": 54800 |
| }, |
| { |
| "epoch": 4.269533827582191, |
| "grad_norm": 2.6232857704162598, |
| "learning_rate": 8.139695354413567e-07, |
| "loss": 0.2193, |
| "step": 55000 |
| }, |
| { |
| "epoch": 4.285059969724022, |
| "grad_norm": 3.286308765411377, |
| "learning_rate": 7.967189359830255e-07, |
| "loss": 0.2454, |
| "step": 55200 |
| }, |
| { |
| "epoch": 4.300586111865854, |
| "grad_norm": 4.827431678771973, |
| "learning_rate": 7.794683365246943e-07, |
| "loss": 0.2417, |
| "step": 55400 |
| }, |
| { |
| "epoch": 4.3161122540076855, |
| "grad_norm": 0.9785335063934326, |
| "learning_rate": 7.622177370663632e-07, |
| "loss": 0.2615, |
| "step": 55600 |
| }, |
| { |
| "epoch": 4.331638396149517, |
| "grad_norm": 3.16576886177063, |
| "learning_rate": 7.44967137608032e-07, |
| "loss": 0.2606, |
| "step": 55800 |
| }, |
| { |
| "epoch": 4.347164538291348, |
| "grad_norm": 1.6270920038223267, |
| "learning_rate": 7.277165381497008e-07, |
| "loss": 0.2608, |
| "step": 56000 |
| }, |
| { |
| "epoch": 4.347164538291348, |
| "eval_loss": 0.09759292006492615, |
| "eval_runtime": 1650.1228, |
| "eval_samples_per_second": 14.681, |
| "eval_steps_per_second": 1.836, |
| "eval_wer": 0.10534638476135173, |
| "step": 56000 |
| }, |
| { |
| "epoch": 4.36269068043318, |
| "grad_norm": 0.7712479829788208, |
| "learning_rate": 7.104659386913696e-07, |
| "loss": 0.2803, |
| "step": 56200 |
| }, |
| { |
| "epoch": 4.378216822575011, |
| "grad_norm": 4.079743385314941, |
| "learning_rate": 6.932153392330385e-07, |
| "loss": 0.2419, |
| "step": 56400 |
| }, |
| { |
| "epoch": 4.393742964716842, |
| "grad_norm": 2.082552671432495, |
| "learning_rate": 6.759647397747073e-07, |
| "loss": 0.2506, |
| "step": 56600 |
| }, |
| { |
| "epoch": 4.409269106858673, |
| "grad_norm": 6.604489803314209, |
| "learning_rate": 6.58714140316376e-07, |
| "loss": 0.2573, |
| "step": 56800 |
| }, |
| { |
| "epoch": 4.424795249000504, |
| "grad_norm": 0.8739509582519531, |
| "learning_rate": 6.414635408580448e-07, |
| "loss": 0.2419, |
| "step": 57000 |
| }, |
| { |
| "epoch": 4.440321391142336, |
| "grad_norm": 3.0703377723693848, |
| "learning_rate": 6.242129413997137e-07, |
| "loss": 0.2423, |
| "step": 57200 |
| }, |
| { |
| "epoch": 4.455847533284167, |
| "grad_norm": 3.4747469425201416, |
| "learning_rate": 6.069623419413825e-07, |
| "loss": 0.2584, |
| "step": 57400 |
| }, |
| { |
| "epoch": 4.4713736754259985, |
| "grad_norm": 4.5395355224609375, |
| "learning_rate": 5.897117424830513e-07, |
| "loss": 0.228, |
| "step": 57600 |
| }, |
| { |
| "epoch": 4.48689981756783, |
| "grad_norm": 1.6975964307785034, |
| "learning_rate": 5.724611430247202e-07, |
| "loss": 0.2485, |
| "step": 57800 |
| }, |
| { |
| "epoch": 4.502425959709662, |
| "grad_norm": 6.7413835525512695, |
| "learning_rate": 5.552105435663889e-07, |
| "loss": 0.2382, |
| "step": 58000 |
| }, |
| { |
| "epoch": 4.502425959709662, |
| "eval_loss": 0.09524416923522949, |
| "eval_runtime": 1476.8186, |
| "eval_samples_per_second": 16.404, |
| "eval_steps_per_second": 2.051, |
| "eval_wer": 0.10636516347120978, |
| "step": 58000 |
| }, |
| { |
| "epoch": 4.517952101851493, |
| "grad_norm": 18.79282569885254, |
| "learning_rate": 5.379599441080578e-07, |
| "loss": 0.2626, |
| "step": 58200 |
| }, |
| { |
| "epoch": 4.533478243993324, |
| "grad_norm": 3.155864953994751, |
| "learning_rate": 5.207093446497266e-07, |
| "loss": 0.2443, |
| "step": 58400 |
| }, |
| { |
| "epoch": 4.549004386135155, |
| "grad_norm": 1.7882168292999268, |
| "learning_rate": 5.034587451913954e-07, |
| "loss": 0.2565, |
| "step": 58600 |
| }, |
| { |
| "epoch": 4.564530528276986, |
| "grad_norm": 4.624693870544434, |
| "learning_rate": 4.862081457330642e-07, |
| "loss": 0.2737, |
| "step": 58800 |
| }, |
| { |
| "epoch": 4.580056670418818, |
| "grad_norm": 8.17938232421875, |
| "learning_rate": 4.689575462747331e-07, |
| "loss": 0.2518, |
| "step": 59000 |
| }, |
| { |
| "epoch": 4.595582812560649, |
| "grad_norm": 10.564002990722656, |
| "learning_rate": 4.517069468164019e-07, |
| "loss": 0.2582, |
| "step": 59200 |
| }, |
| { |
| "epoch": 4.61110895470248, |
| "grad_norm": 4.8782477378845215, |
| "learning_rate": 4.3445634735807076e-07, |
| "loss": 0.2724, |
| "step": 59400 |
| }, |
| { |
| "epoch": 4.6266350968443115, |
| "grad_norm": 3.1332931518554688, |
| "learning_rate": 4.1720574789973956e-07, |
| "loss": 0.2292, |
| "step": 59600 |
| }, |
| { |
| "epoch": 4.642161238986143, |
| "grad_norm": 1.7788443565368652, |
| "learning_rate": 4.0004140143870004e-07, |
| "loss": 0.2538, |
| "step": 59800 |
| }, |
| { |
| "epoch": 4.657687381127975, |
| "grad_norm": 6.25827693939209, |
| "learning_rate": 3.827908019803688e-07, |
| "loss": 0.2657, |
| "step": 60000 |
| }, |
| { |
| "epoch": 4.657687381127975, |
| "eval_loss": 0.09603731334209442, |
| "eval_runtime": 1478.0921, |
| "eval_samples_per_second": 16.389, |
| "eval_steps_per_second": 2.049, |
| "eval_wer": 0.10629776335300034, |
| "step": 60000 |
| }, |
| { |
| "epoch": 4.673213523269806, |
| "grad_norm": 4.697340488433838, |
| "learning_rate": 3.6554020252203765e-07, |
| "loss": 0.254, |
| "step": 60200 |
| }, |
| { |
| "epoch": 4.688739665411637, |
| "grad_norm": 2.947510004043579, |
| "learning_rate": 3.4828960306370645e-07, |
| "loss": 0.2363, |
| "step": 60400 |
| }, |
| { |
| "epoch": 4.704265807553468, |
| "grad_norm": 4.655181884765625, |
| "learning_rate": 3.310390036053753e-07, |
| "loss": 0.2515, |
| "step": 60600 |
| }, |
| { |
| "epoch": 4.719791949695299, |
| "grad_norm": 2.0875043869018555, |
| "learning_rate": 3.137884041470441e-07, |
| "loss": 0.2516, |
| "step": 60800 |
| }, |
| { |
| "epoch": 4.735318091837131, |
| "grad_norm": 6.2482428550720215, |
| "learning_rate": 2.9653780468871297e-07, |
| "loss": 0.2764, |
| "step": 61000 |
| }, |
| { |
| "epoch": 4.750844233978962, |
| "grad_norm": 3.430018901824951, |
| "learning_rate": 2.7928720523038177e-07, |
| "loss": 0.2283, |
| "step": 61200 |
| }, |
| { |
| "epoch": 4.766370376120793, |
| "grad_norm": 1.6706165075302124, |
| "learning_rate": 2.6203660577205063e-07, |
| "loss": 0.2611, |
| "step": 61400 |
| }, |
| { |
| "epoch": 4.7818965182626245, |
| "grad_norm": 4.43579626083374, |
| "learning_rate": 2.4478600631371943e-07, |
| "loss": 0.2561, |
| "step": 61600 |
| }, |
| { |
| "epoch": 4.797422660404456, |
| "grad_norm": 3.309246778488159, |
| "learning_rate": 2.2753540685538824e-07, |
| "loss": 0.2802, |
| "step": 61800 |
| }, |
| { |
| "epoch": 4.812948802546288, |
| "grad_norm": 1.4542226791381836, |
| "learning_rate": 2.1028480739705706e-07, |
| "loss": 0.252, |
| "step": 62000 |
| }, |
| { |
| "epoch": 4.812948802546288, |
| "eval_loss": 0.09494243562221527, |
| "eval_runtime": 1450.0602, |
| "eval_samples_per_second": 16.706, |
| "eval_steps_per_second": 2.089, |
| "eval_wer": 0.10604112444135672, |
| "step": 62000 |
| }, |
| { |
| "epoch": 4.828474944688119, |
| "grad_norm": 2.262542724609375, |
| "learning_rate": 1.930342079387259e-07, |
| "loss": 0.2585, |
| "step": 62200 |
| }, |
| { |
| "epoch": 4.84400108682995, |
| "grad_norm": 1.6216250658035278, |
| "learning_rate": 1.7578360848039472e-07, |
| "loss": 0.2337, |
| "step": 62400 |
| }, |
| { |
| "epoch": 4.859527228971781, |
| "grad_norm": 1.9740136861801147, |
| "learning_rate": 1.5853300902206353e-07, |
| "loss": 0.2363, |
| "step": 62600 |
| }, |
| { |
| "epoch": 4.875053371113612, |
| "grad_norm": 2.8920953273773193, |
| "learning_rate": 1.4128240956373236e-07, |
| "loss": 0.2507, |
| "step": 62800 |
| }, |
| { |
| "epoch": 4.890579513255444, |
| "grad_norm": 3.6626710891723633, |
| "learning_rate": 1.240318101054012e-07, |
| "loss": 0.2576, |
| "step": 63000 |
| }, |
| { |
| "epoch": 4.906105655397275, |
| "grad_norm": 4.759396553039551, |
| "learning_rate": 1.0678121064707e-07, |
| "loss": 0.2421, |
| "step": 63200 |
| }, |
| { |
| "epoch": 4.921631797539106, |
| "grad_norm": 1.2484747171401978, |
| "learning_rate": 8.953061118873882e-08, |
| "loss": 0.2435, |
| "step": 63400 |
| }, |
| { |
| "epoch": 4.9371579396809375, |
| "grad_norm": 3.4302594661712646, |
| "learning_rate": 7.228001173040764e-08, |
| "loss": 0.2518, |
| "step": 63600 |
| }, |
| { |
| "epoch": 4.952684081822769, |
| "grad_norm": 4.041806697845459, |
| "learning_rate": 5.502941227207646e-08, |
| "loss": 0.2497, |
| "step": 63800 |
| }, |
| { |
| "epoch": 4.968210223964601, |
| "grad_norm": 1.5748398303985596, |
| "learning_rate": 3.786506581103694e-08, |
| "loss": 0.2622, |
| "step": 64000 |
| }, |
| { |
| "epoch": 4.968210223964601, |
| "eval_loss": 0.09495660662651062, |
| "eval_runtime": 1441.498, |
| "eval_samples_per_second": 16.805, |
| "eval_steps_per_second": 2.101, |
| "eval_wer": 0.1054267464407553, |
| "step": 64000 |
| } |
| ], |
| "logging_steps": 200, |
| "max_steps": 64410, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 2000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 7.172721657986751e+20, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
