| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 198.62941847206386, | |
| "global_step": 87000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 5.988000000000001e-05, | |
| "loss": 1.7702, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_cer": 0.8902483010528606, | |
| "eval_loss": 1.379656434059143, | |
| "eval_runtime": 157.8178, | |
| "eval_samples_per_second": 23.134, | |
| "eval_steps_per_second": 1.451, | |
| "eval_wer": 0.9197613252728272, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 7.906822429906543e-05, | |
| "loss": 0.6423, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_cer": 0.8909467730023416, | |
| "eval_loss": 1.096530795097351, | |
| "eval_runtime": 155.6392, | |
| "eval_samples_per_second": 23.458, | |
| "eval_steps_per_second": 1.471, | |
| "eval_wer": 0.92135398557633, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 7.766728971962618e-05, | |
| "loss": 0.5752, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "eval_cer": 0.8896414319820001, | |
| "eval_loss": 0.9504674077033997, | |
| "eval_runtime": 156.6885, | |
| "eval_samples_per_second": 23.301, | |
| "eval_steps_per_second": 1.461, | |
| "eval_wer": 0.9201763142251483, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 7.626728971962617e-05, | |
| "loss": 0.5136, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "eval_cer": 0.8656071267039567, | |
| "eval_loss": 0.49109727144241333, | |
| "eval_runtime": 157.3097, | |
| "eval_samples_per_second": 23.209, | |
| "eval_steps_per_second": 1.456, | |
| "eval_wer": 0.8955349431913772, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 17.12, | |
| "learning_rate": 7.486542056074767e-05, | |
| "loss": 0.5015, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 17.12, | |
| "eval_cer": 0.818803780908815, | |
| "eval_loss": 0.4844619333744049, | |
| "eval_runtime": 157.3062, | |
| "eval_samples_per_second": 23.21, | |
| "eval_steps_per_second": 1.456, | |
| "eval_wer": 0.849381442142689, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "learning_rate": 7.346355140186916e-05, | |
| "loss": 0.462, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "eval_cer": 0.8027961778698895, | |
| "eval_loss": 0.4619074761867523, | |
| "eval_runtime": 157.0856, | |
| "eval_samples_per_second": 23.242, | |
| "eval_steps_per_second": 1.458, | |
| "eval_wer": 0.8343072488475645, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 23.97, | |
| "learning_rate": 7.206168224299066e-05, | |
| "loss": 0.444, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 23.97, | |
| "eval_cer": 0.7654622796521381, | |
| "eval_loss": 0.4942396581172943, | |
| "eval_runtime": 156.8664, | |
| "eval_samples_per_second": 23.275, | |
| "eval_steps_per_second": 1.46, | |
| "eval_wer": 0.7982929373366682, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "learning_rate": 7.066074766355141e-05, | |
| "loss": 0.4318, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "eval_cer": 0.7344947814985086, | |
| "eval_loss": 0.4564968943595886, | |
| "eval_runtime": 156.676, | |
| "eval_samples_per_second": 23.303, | |
| "eval_steps_per_second": 1.462, | |
| "eval_wer": 0.7668771520541953, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 30.82, | |
| "learning_rate": 6.92588785046729e-05, | |
| "loss": 0.3877, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 30.82, | |
| "eval_cer": 0.7305844836173977, | |
| "eval_loss": 0.44864970445632935, | |
| "eval_runtime": 156.7014, | |
| "eval_samples_per_second": 23.299, | |
| "eval_steps_per_second": 1.461, | |
| "eval_wer": 0.7626823988604627, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 34.25, | |
| "learning_rate": 6.78570093457944e-05, | |
| "loss": 0.3734, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 34.25, | |
| "eval_cer": 0.716420388510709, | |
| "eval_loss": 0.4351600408554077, | |
| "eval_runtime": 156.7512, | |
| "eval_samples_per_second": 23.292, | |
| "eval_steps_per_second": 1.461, | |
| "eval_wer": 0.7498626050090288, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 37.67, | |
| "learning_rate": 6.64551401869159e-05, | |
| "loss": 0.3595, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 37.67, | |
| "eval_cer": 0.6001820607212581, | |
| "eval_loss": 0.4632013738155365, | |
| "eval_runtime": 156.8539, | |
| "eval_samples_per_second": 23.276, | |
| "eval_steps_per_second": 1.46, | |
| "eval_wer": 0.6390493388216557, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "learning_rate": 6.505327102803739e-05, | |
| "loss": 0.3466, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "eval_cer": 0.592630548414984, | |
| "eval_loss": 0.4672980308532715, | |
| "eval_runtime": 155.7423, | |
| "eval_samples_per_second": 23.443, | |
| "eval_steps_per_second": 1.47, | |
| "eval_wer": 0.6305925369284088, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 44.52, | |
| "learning_rate": 6.365140186915889e-05, | |
| "loss": 0.3346, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 44.52, | |
| "eval_cer": 0.5607641970148912, | |
| "eval_loss": 0.48392337560653687, | |
| "eval_runtime": 156.8128, | |
| "eval_samples_per_second": 23.283, | |
| "eval_steps_per_second": 1.46, | |
| "eval_wer": 0.601790060453796, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "learning_rate": 6.224953271028038e-05, | |
| "loss": 0.3224, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "eval_cer": 0.5324932586006516, | |
| "eval_loss": 0.47312837839126587, | |
| "eval_runtime": 155.7519, | |
| "eval_samples_per_second": 23.441, | |
| "eval_steps_per_second": 1.47, | |
| "eval_wer": 0.5755448131988918, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 51.37, | |
| "learning_rate": 6.0849532710280374e-05, | |
| "loss": 0.3115, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 51.37, | |
| "eval_cer": 0.5504359724504343, | |
| "eval_loss": 0.46229076385498047, | |
| "eval_runtime": 157.1658, | |
| "eval_samples_per_second": 23.23, | |
| "eval_steps_per_second": 1.457, | |
| "eval_wer": 0.5916060072454827, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 54.79, | |
| "learning_rate": 5.9448598130841125e-05, | |
| "loss": 0.3004, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 54.79, | |
| "eval_cer": 0.4698426148041702, | |
| "eval_loss": 0.4939606785774231, | |
| "eval_runtime": 157.0777, | |
| "eval_samples_per_second": 23.243, | |
| "eval_steps_per_second": 1.458, | |
| "eval_wer": 0.5165603023811393, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 58.22, | |
| "learning_rate": 5.804672897196262e-05, | |
| "loss": 0.2911, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 58.22, | |
| "eval_cer": 0.4784017587752695, | |
| "eval_loss": 0.49099990725517273, | |
| "eval_runtime": 155.9127, | |
| "eval_samples_per_second": 23.417, | |
| "eval_steps_per_second": 1.469, | |
| "eval_wer": 0.5228972958422593, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 61.64, | |
| "learning_rate": 5.664579439252337e-05, | |
| "loss": 0.2816, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 61.64, | |
| "eval_cer": 0.48053152570319524, | |
| "eval_loss": 0.48187437653541565, | |
| "eval_runtime": 157.1059, | |
| "eval_samples_per_second": 23.239, | |
| "eval_steps_per_second": 1.458, | |
| "eval_wer": 0.524153478616853, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 65.07, | |
| "learning_rate": 5.5244859813084115e-05, | |
| "loss": 0.2753, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 65.07, | |
| "eval_cer": 0.43459840725494797, | |
| "eval_loss": 0.5083651542663574, | |
| "eval_runtime": 156.7879, | |
| "eval_samples_per_second": 23.286, | |
| "eval_steps_per_second": 1.461, | |
| "eval_wer": 0.48194798057403065, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 68.49, | |
| "learning_rate": 5.384299065420561e-05, | |
| "loss": 0.2628, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 68.49, | |
| "eval_cer": 0.441067860557518, | |
| "eval_loss": 0.5106444954872131, | |
| "eval_runtime": 161.9652, | |
| "eval_samples_per_second": 22.542, | |
| "eval_steps_per_second": 1.414, | |
| "eval_wer": 0.4881952466941083, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 71.92, | |
| "learning_rate": 5.2441121495327106e-05, | |
| "loss": 0.2534, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 71.92, | |
| "eval_cer": 0.40500495228062544, | |
| "eval_loss": 0.555060088634491, | |
| "eval_runtime": 162.1242, | |
| "eval_samples_per_second": 22.52, | |
| "eval_steps_per_second": 1.412, | |
| "eval_wer": 0.45390818649827835, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 75.34, | |
| "learning_rate": 5.104018691588786e-05, | |
| "loss": 0.2464, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 75.34, | |
| "eval_cer": 0.4066767048154488, | |
| "eval_loss": 0.5396202206611633, | |
| "eval_runtime": 161.0534, | |
| "eval_samples_per_second": 22.669, | |
| "eval_steps_per_second": 1.422, | |
| "eval_wer": 0.45515315335524176, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 78.77, | |
| "learning_rate": 4.96392523364486e-05, | |
| "loss": 0.2447, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 78.77, | |
| "eval_cer": 0.38928360823738883, | |
| "eval_loss": 0.5427628755569458, | |
| "eval_runtime": 161.5749, | |
| "eval_samples_per_second": 22.596, | |
| "eval_steps_per_second": 1.417, | |
| "eval_wer": 0.4384190042508328, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 82.19, | |
| "learning_rate": 4.82392523364486e-05, | |
| "loss": 0.2311, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 82.19, | |
| "eval_cer": 0.3919687176169511, | |
| "eval_loss": 0.537956714630127, | |
| "eval_runtime": 161.1887, | |
| "eval_samples_per_second": 22.65, | |
| "eval_steps_per_second": 1.421, | |
| "eval_wer": 0.43954059601386286, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 85.62, | |
| "learning_rate": 4.6837383177570096e-05, | |
| "loss": 0.2238, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 85.62, | |
| "eval_cer": 0.36581037059089583, | |
| "eval_loss": 0.5633525848388672, | |
| "eval_runtime": 160.6641, | |
| "eval_samples_per_second": 22.724, | |
| "eval_steps_per_second": 1.425, | |
| "eval_wer": 0.4180508978342063, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 89.04, | |
| "learning_rate": 4.543644859813085e-05, | |
| "loss": 0.2176, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 89.04, | |
| "eval_cer": 0.34971688985326366, | |
| "eval_loss": 0.5895515084266663, | |
| "eval_runtime": 161.4249, | |
| "eval_samples_per_second": 22.617, | |
| "eval_steps_per_second": 1.419, | |
| "eval_wer": 0.4021355107168093, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 92.47, | |
| "learning_rate": 4.4034579439252336e-05, | |
| "loss": 0.2106, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 92.47, | |
| "eval_cer": 0.354417262562476, | |
| "eval_loss": 0.5802760124206543, | |
| "eval_runtime": 161.093, | |
| "eval_samples_per_second": 22.664, | |
| "eval_steps_per_second": 1.422, | |
| "eval_wer": 0.40675646878049326, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 95.89, | |
| "learning_rate": 4.263271028037384e-05, | |
| "loss": 0.2058, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 95.89, | |
| "eval_cer": 0.34699742939421874, | |
| "eval_loss": 0.5864582657814026, | |
| "eval_runtime": 161.933, | |
| "eval_samples_per_second": 22.546, | |
| "eval_steps_per_second": 1.414, | |
| "eval_wer": 0.400744736930652, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 99.31, | |
| "learning_rate": 4.1230841121495333e-05, | |
| "loss": 0.1995, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 99.31, | |
| "eval_cer": 0.33578752712303983, | |
| "eval_loss": 0.5888575315475464, | |
| "eval_runtime": 161.5039, | |
| "eval_samples_per_second": 22.606, | |
| "eval_steps_per_second": 1.418, | |
| "eval_wer": 0.3900335355937146, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 102.74, | |
| "learning_rate": 3.982990654205608e-05, | |
| "loss": 0.1956, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 102.74, | |
| "eval_cer": 0.3373562264194152, | |
| "eval_loss": 0.5785849094390869, | |
| "eval_runtime": 159.8624, | |
| "eval_samples_per_second": 22.838, | |
| "eval_steps_per_second": 1.432, | |
| "eval_wer": 0.3900335355937146, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 106.16, | |
| "learning_rate": 3.842803738317757e-05, | |
| "loss": 0.191, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 106.16, | |
| "eval_cer": 0.33468256739968055, | |
| "eval_loss": 0.5981589555740356, | |
| "eval_runtime": 159.9077, | |
| "eval_samples_per_second": 22.832, | |
| "eval_steps_per_second": 1.432, | |
| "eval_wer": 0.38939422828878745, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 109.59, | |
| "learning_rate": 3.7027102803738324e-05, | |
| "loss": 0.1851, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 109.59, | |
| "eval_cer": 0.32407380901944843, | |
| "eval_loss": 0.603424072265625, | |
| "eval_runtime": 159.431, | |
| "eval_samples_per_second": 22.9, | |
| "eval_steps_per_second": 1.436, | |
| "eval_wer": 0.3804551419374376, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 113.01, | |
| "learning_rate": 3.562616822429907e-05, | |
| "loss": 0.1822, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 113.01, | |
| "eval_cer": 0.30877612828983153, | |
| "eval_loss": 0.6061017513275146, | |
| "eval_runtime": 158.5159, | |
| "eval_samples_per_second": 23.032, | |
| "eval_steps_per_second": 1.445, | |
| "eval_wer": 0.3677250754270461, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 116.44, | |
| "learning_rate": 3.422523364485982e-05, | |
| "loss": 0.183, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 116.44, | |
| "eval_cer": 0.3065089570439751, | |
| "eval_loss": 0.6075730919837952, | |
| "eval_runtime": 160.2872, | |
| "eval_samples_per_second": 22.778, | |
| "eval_steps_per_second": 1.429, | |
| "eval_wer": 0.36392287935037404, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 119.86, | |
| "learning_rate": 3.2823364485981314e-05, | |
| "loss": 0.172, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 119.86, | |
| "eval_cer": 0.2992551540932174, | |
| "eval_loss": 0.6227861642837524, | |
| "eval_runtime": 160.5993, | |
| "eval_samples_per_second": 22.734, | |
| "eval_steps_per_second": 1.426, | |
| "eval_wer": 0.3576083177245146, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 123.29, | |
| "learning_rate": 3.14214953271028e-05, | |
| "loss": 0.1677, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 123.29, | |
| "eval_cer": 0.300377289356318, | |
| "eval_loss": 0.630536675453186, | |
| "eval_runtime": 158.8853, | |
| "eval_samples_per_second": 22.979, | |
| "eval_steps_per_second": 1.441, | |
| "eval_wer": 0.35942529638062337, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 126.71, | |
| "learning_rate": 3.002056074766355e-05, | |
| "loss": 0.1641, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 126.71, | |
| "eval_cer": 0.2999479008627846, | |
| "eval_loss": 0.6270943284034729, | |
| "eval_runtime": 160.9314, | |
| "eval_samples_per_second": 22.687, | |
| "eval_steps_per_second": 1.423, | |
| "eval_wer": 0.3598515012505748, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 130.14, | |
| "learning_rate": 2.861869158878505e-05, | |
| "loss": 0.1596, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 130.14, | |
| "eval_cer": 0.3011845397241608, | |
| "eval_loss": 0.6379602551460266, | |
| "eval_runtime": 160.483, | |
| "eval_samples_per_second": 22.75, | |
| "eval_steps_per_second": 1.427, | |
| "eval_wer": 0.3607712064962595, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 133.56, | |
| "learning_rate": 2.7217757009345797e-05, | |
| "loss": 0.1561, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 133.56, | |
| "eval_cer": 0.2984937051646848, | |
| "eval_loss": 0.6372924447059631, | |
| "eval_runtime": 149.4703, | |
| "eval_samples_per_second": 24.426, | |
| "eval_steps_per_second": 1.532, | |
| "eval_wer": 0.3564530782085936, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 136.99, | |
| "learning_rate": 2.5815887850467292e-05, | |
| "loss": 0.1531, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 136.99, | |
| "eval_cer": 0.29037540004694645, | |
| "eval_loss": 0.633769154548645, | |
| "eval_runtime": 149.3294, | |
| "eval_samples_per_second": 24.449, | |
| "eval_steps_per_second": 1.534, | |
| "eval_wer": 0.35054228961742506, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 140.41, | |
| "learning_rate": 2.441495327102804e-05, | |
| "loss": 0.1501, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 140.41, | |
| "eval_cer": 0.280087251741886, | |
| "eval_loss": 0.6684455275535583, | |
| "eval_runtime": 148.2869, | |
| "eval_samples_per_second": 24.621, | |
| "eval_steps_per_second": 1.544, | |
| "eval_wer": 0.34238831750019627, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 143.83, | |
| "learning_rate": 2.3013084112149535e-05, | |
| "loss": 0.1461, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 143.83, | |
| "eval_cer": 0.2885776935540199, | |
| "eval_loss": 0.653037428855896, | |
| "eval_runtime": 148.0728, | |
| "eval_samples_per_second": 24.657, | |
| "eval_steps_per_second": 1.547, | |
| "eval_wer": 0.34928610684283135, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 147.26, | |
| "learning_rate": 2.1612149532710283e-05, | |
| "loss": 0.1438, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 147.26, | |
| "eval_cer": 0.28241739996679394, | |
| "eval_loss": 0.6683825254440308, | |
| "eval_runtime": 149.486, | |
| "eval_samples_per_second": 24.424, | |
| "eval_steps_per_second": 1.532, | |
| "eval_wer": 0.3449006830493837, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 150.68, | |
| "learning_rate": 2.0210280373831778e-05, | |
| "loss": 0.1399, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 150.68, | |
| "eval_cer": 0.2820109121929157, | |
| "eval_loss": 0.6699740290641785, | |
| "eval_runtime": 148.9217, | |
| "eval_samples_per_second": 24.516, | |
| "eval_steps_per_second": 1.538, | |
| "eval_wer": 0.34547269484852905, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 154.11, | |
| "learning_rate": 1.8810280373831778e-05, | |
| "loss": 0.1384, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 154.11, | |
| "eval_cer": 0.2813067150635209, | |
| "eval_loss": 0.6824755668640137, | |
| "eval_runtime": 149.3793, | |
| "eval_samples_per_second": 24.441, | |
| "eval_steps_per_second": 1.533, | |
| "eval_wer": 0.34288181787592953, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 157.53, | |
| "learning_rate": 1.740841121495327e-05, | |
| "loss": 0.1356, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 157.53, | |
| "eval_cer": 0.28230862154843217, | |
| "eval_loss": 0.6769393086433411, | |
| "eval_runtime": 147.7768, | |
| "eval_samples_per_second": 24.706, | |
| "eval_steps_per_second": 1.55, | |
| "eval_wer": 0.34468758061440796, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 160.96, | |
| "learning_rate": 1.600654205607477e-05, | |
| "loss": 0.1327, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 160.96, | |
| "eval_cer": 0.2698334545163082, | |
| "eval_loss": 0.6889060735702515, | |
| "eval_runtime": 148.0859, | |
| "eval_samples_per_second": 24.655, | |
| "eval_steps_per_second": 1.546, | |
| "eval_wer": 0.3343016408887493, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 164.38, | |
| "learning_rate": 1.4604672897196264e-05, | |
| "loss": 0.1301, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 164.38, | |
| "eval_cer": 0.2741445149913836, | |
| "eval_loss": 0.667121171951294, | |
| "eval_runtime": 147.8905, | |
| "eval_samples_per_second": 24.687, | |
| "eval_steps_per_second": 1.548, | |
| "eval_wer": 0.33721777947262754, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 167.81, | |
| "learning_rate": 1.3204672897196262e-05, | |
| "loss": 0.1274, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 167.81, | |
| "eval_cer": 0.2703658962482896, | |
| "eval_loss": 0.6766042113304138, | |
| "eval_runtime": 157.2135, | |
| "eval_samples_per_second": 23.223, | |
| "eval_steps_per_second": 1.457, | |
| "eval_wer": 0.33454839107661594, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 171.23, | |
| "learning_rate": 1.180373831775701e-05, | |
| "loss": 0.1254, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 171.23, | |
| "eval_cer": 0.26700521563890145, | |
| "eval_loss": 0.6990349888801575, | |
| "eval_runtime": 147.5703, | |
| "eval_samples_per_second": 24.741, | |
| "eval_steps_per_second": 1.552, | |
| "eval_wer": 0.33097051335254996, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 174.66, | |
| "learning_rate": 1.0401869158878507e-05, | |
| "loss": 0.1232, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 174.66, | |
| "eval_cer": 0.26890025019036223, | |
| "eval_loss": 0.6996525526046753, | |
| "eval_runtime": 148.6819, | |
| "eval_samples_per_second": 24.556, | |
| "eval_steps_per_second": 1.54, | |
| "eval_wer": 0.3331015377023071, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 178.08, | |
| "learning_rate": 9e-06, | |
| "loss": 0.1214, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 178.08, | |
| "eval_cer": 0.2654765926019225, | |
| "eval_loss": 0.7195149660110474, | |
| "eval_runtime": 147.4964, | |
| "eval_samples_per_second": 24.753, | |
| "eval_steps_per_second": 1.553, | |
| "eval_wer": 0.33113875211700444, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 181.51, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.1192, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 181.51, | |
| "eval_cer": 0.2665872775051956, | |
| "eval_loss": 0.7197592854499817, | |
| "eval_runtime": 146.5503, | |
| "eval_samples_per_second": 24.913, | |
| "eval_steps_per_second": 1.563, | |
| "eval_wer": 0.3315873888222165, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 184.93, | |
| "learning_rate": 6.198130841121495e-06, | |
| "loss": 0.117, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 184.93, | |
| "eval_cer": 0.26343270337270347, | |
| "eval_loss": 0.7252310514450073, | |
| "eval_runtime": 147.1458, | |
| "eval_samples_per_second": 24.812, | |
| "eval_steps_per_second": 1.556, | |
| "eval_wer": 0.3294339326371987, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 188.36, | |
| "learning_rate": 4.796261682242991e-06, | |
| "loss": 0.1155, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 188.36, | |
| "eval_cer": 0.26273995660313626, | |
| "eval_loss": 0.7319917678833008, | |
| "eval_runtime": 145.3615, | |
| "eval_samples_per_second": 25.117, | |
| "eval_steps_per_second": 1.575, | |
| "eval_wer": 0.3283684204623201, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 191.78, | |
| "learning_rate": 3.3953271028037387e-06, | |
| "loss": 0.1146, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 191.78, | |
| "eval_cer": 0.2623048429296891, | |
| "eval_loss": 0.7414848208427429, | |
| "eval_runtime": 146.2983, | |
| "eval_samples_per_second": 24.956, | |
| "eval_steps_per_second": 1.565, | |
| "eval_wer": 0.32897408001435635, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 195.21, | |
| "learning_rate": 1.9934579439252336e-06, | |
| "loss": 0.1131, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 195.21, | |
| "eval_cer": 0.26136018824391555, | |
| "eval_loss": 0.7424480319023132, | |
| "eval_runtime": 146.0724, | |
| "eval_samples_per_second": 24.994, | |
| "eval_steps_per_second": 1.568, | |
| "eval_wer": 0.328177749862605, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 198.63, | |
| "learning_rate": 5.925233644859814e-07, | |
| "loss": 0.1127, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 198.63, | |
| "eval_cer": 0.261194158026416, | |
| "eval_loss": 0.7422526478767395, | |
| "eval_runtime": 146.2208, | |
| "eval_samples_per_second": 24.969, | |
| "eval_steps_per_second": 1.566, | |
| "eval_wer": 0.3272580446169203, | |
| "step": 87000 | |
| } | |
| ], | |
| "max_steps": 87600, | |
| "num_train_epochs": 200, | |
| "total_flos": 9.618016300986444e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |