| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.936550491510276, | |
| "eval_steps": 200, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0467, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.994863481875841e-05, | |
| "loss": 0.0433, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_f1_score": 0.6251468860164512, | |
| "eval_label_f1": 0.8319623971797884, | |
| "eval_loss": 0.05231842026114464, | |
| "eval_runtime": 457.2641, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.10434244136543076, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.979475034558115e-05, | |
| "loss": 0.0402, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.9538978924776634e-05, | |
| "loss": 0.0391, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_f1_score": 0.6206896551724138, | |
| "eval_label_f1": 0.8346394984326019, | |
| "eval_loss": 0.050447478890419006, | |
| "eval_runtime": 457.7184, | |
| "eval_samples_per_second": 2.185, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.10472946822509482, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.9182371575975736e-05, | |
| "loss": 0.037, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.8726393675266716e-05, | |
| "loss": 0.0381, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_f1_score": 0.6142467886337096, | |
| "eval_label_f1": 0.8322304398598676, | |
| "eval_loss": 0.049576789140701294, | |
| "eval_runtime": 458.238, | |
| "eval_samples_per_second": 2.182, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.10654849446551591, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.817291893365055e-05, | |
| "loss": 0.037, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.0374, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_f1_score": 0.6158081006685017, | |
| "eval_label_f1": 0.8360204482894219, | |
| "eval_loss": 0.04838084056973457, | |
| "eval_runtime": 456.7741, | |
| "eval_samples_per_second": 2.189, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.10705162938307919, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 4.678296760308474e-05, | |
| "loss": 0.0375, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 4.595220262229601e-05, | |
| "loss": 0.0374, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_f1_score": 0.6155063291139241, | |
| "eval_label_f1": 0.8370253164556962, | |
| "eval_loss": 0.047435563057661057, | |
| "eval_runtime": 463.3137, | |
| "eval_samples_per_second": 2.158, | |
| "eval_steps_per_second": 0.54, | |
| "eval_wer": 0.10693552132517997, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 4.503534054669892e-05, | |
| "loss": 0.0379, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.4036148959228365e-05, | |
| "loss": 0.0342, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_f1_score": 0.6118110236220473, | |
| "eval_label_f1": 0.8362204724409448, | |
| "eval_loss": 0.04737536609172821, | |
| "eval_runtime": 458.8739, | |
| "eval_samples_per_second": 2.179, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 0.1077095750445081, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.2958733752443195e-05, | |
| "loss": 0.0345, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.180752225653292e-05, | |
| "loss": 0.0362, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_f1_score": 0.6138147566718996, | |
| "eval_label_f1": 0.8375196232339089, | |
| "eval_loss": 0.04683598503470421, | |
| "eval_runtime": 458.5195, | |
| "eval_samples_per_second": 2.181, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 0.10790308847434012, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 4.058724504646834e-05, | |
| "loss": 0.0343, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 3.9302916503054246e-05, | |
| "loss": 0.0351, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_f1_score": 0.6101960784313726, | |
| "eval_label_f1": 0.836078431372549, | |
| "eval_loss": 0.04606114700436592, | |
| "eval_runtime": 460.3927, | |
| "eval_samples_per_second": 2.172, | |
| "eval_steps_per_second": 0.543, | |
| "eval_wer": 0.10821270996207137, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 3.7959814207763135e-05, | |
| "loss": 0.0355, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 3.656345725602089e-05, | |
| "loss": 0.0339, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_f1_score": 0.611111111111111, | |
| "eval_label_f1": 0.838810641627543, | |
| "eval_loss": 0.046606115996837616, | |
| "eval_runtime": 460.841, | |
| "eval_samples_per_second": 2.17, | |
| "eval_steps_per_second": 0.542, | |
| "eval_wer": 0.10794179116030653, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3.5119583578059846e-05, | |
| "loss": 0.0333, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 3.363412636053269e-05, | |
| "loss": 0.0323, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_f1_score": 0.6168371361132967, | |
| "eval_label_f1": 0.8418568056648308, | |
| "eval_loss": 0.04674151912331581, | |
| "eval_runtime": 460.8468, | |
| "eval_samples_per_second": 2.17, | |
| "eval_steps_per_second": 0.542, | |
| "eval_wer": 0.10879325025156746, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 3.211318966577581e-05, | |
| "loss": 0.0332, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.056302334890786e-05, | |
| "loss": 0.0338, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "eval_f1_score": 0.6093443266588143, | |
| "eval_label_f1": 0.8425598743619945, | |
| "eval_loss": 0.0457298718392849, | |
| "eval_runtime": 456.4605, | |
| "eval_samples_per_second": 2.191, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.10856103413576902, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 2.8989997375834482e-05, | |
| "loss": 0.0333, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 2.7400575647692046e-05, | |
| "loss": 0.032, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "eval_f1_score": 0.6090404440919905, | |
| "eval_label_f1": 0.8398096748612214, | |
| "eval_loss": 0.045234858989715576, | |
| "eval_runtime": 457.189, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.10852233144980261, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 2.5801289439291388e-05, | |
| "loss": 0.0335, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 2.419871056070862e-05, | |
| "loss": 0.0307, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "eval_f1_score": 0.6139315230224321, | |
| "eval_label_f1": 0.8421881149153877, | |
| "eval_loss": 0.04505770280957222, | |
| "eval_runtime": 456.5336, | |
| "eval_samples_per_second": 2.19, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.10859973682173543, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 2.2599424352307957e-05, | |
| "loss": 0.0324, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.1010002624165527e-05, | |
| "loss": 0.0321, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1_score": 0.6115702479338844, | |
| "eval_label_f1": 0.8398268398268398, | |
| "eval_loss": 0.04516015574336052, | |
| "eval_runtime": 459.143, | |
| "eval_samples_per_second": 2.178, | |
| "eval_steps_per_second": 0.544, | |
| "eval_wer": 0.10829011533400418, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 0.0303, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 1.7886810334224192e-05, | |
| "loss": 0.0313, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "eval_f1_score": 0.6116352201257862, | |
| "eval_label_f1": 0.8404088050314464, | |
| "eval_loss": 0.044775452464818954, | |
| "eval_runtime": 458.1981, | |
| "eval_samples_per_second": 2.182, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.10921897979719793, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 1.6365873639467315e-05, | |
| "loss": 0.0321, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 1.4880416421940155e-05, | |
| "loss": 0.0309, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "eval_f1_score": 0.6109148017275227, | |
| "eval_label_f1": 0.8402041617589321, | |
| "eval_loss": 0.04491139575839043, | |
| "eval_runtime": 457.2993, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.10832881801997059, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 1.3436542743979125e-05, | |
| "loss": 0.0318, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 1.2040185792236874e-05, | |
| "loss": 0.0305, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "eval_f1_score": 0.6085590891244601, | |
| "eval_label_f1": 0.8402041617589321, | |
| "eval_loss": 0.0448303148150444, | |
| "eval_runtime": 459.7209, | |
| "eval_samples_per_second": 2.175, | |
| "eval_steps_per_second": 0.544, | |
| "eval_wer": 0.10825141264803777, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 1.0697083496945765e-05, | |
| "loss": 0.0302, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 9.412754953531663e-06, | |
| "loss": 0.0301, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "eval_f1_score": 0.6115702479338844, | |
| "eval_label_f1": 0.8374655647382919, | |
| "eval_loss": 0.044706329703330994, | |
| "eval_runtime": 459.3177, | |
| "eval_samples_per_second": 2.177, | |
| "eval_steps_per_second": 0.544, | |
| "eval_wer": 0.10813530459013855, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 8.192477743467078e-06, | |
| "loss": 0.0304, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 7.041266247556813e-06, | |
| "loss": 0.03, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "eval_f1_score": 0.6103434662455587, | |
| "eval_label_f1": 0.8401105408606395, | |
| "eval_loss": 0.04461517930030823, | |
| "eval_runtime": 456.0173, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.10867714219366824, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 5.9638510407716394e-06, | |
| "loss": 0.0314, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 4.9646594533010875e-06, | |
| "loss": 0.0302, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "eval_f1_score": 0.6120047077285209, | |
| "eval_label_f1": 0.8387602981561396, | |
| "eval_loss": 0.04448845237493515, | |
| "eval_runtime": 457.6573, | |
| "eval_samples_per_second": 2.185, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.108367520705937, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 4.047797377703985e-06, | |
| "loss": 0.0309, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 3.217032396915265e-06, | |
| "loss": 0.0294, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "eval_f1_score": 0.6132075471698112, | |
| "eval_label_f1": 0.8396226415094339, | |
| "eval_loss": 0.044213637709617615, | |
| "eval_runtime": 455.3785, | |
| "eval_samples_per_second": 2.196, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 0.10863843950770183, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 2.475778302439524e-06, | |
| "loss": 0.0296, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 1.827081066349459e-06, | |
| "loss": 0.03, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "eval_f1_score": 0.6111547525530244, | |
| "eval_label_f1": 0.8381775333857032, | |
| "eval_loss": 0.04440901055932045, | |
| "eval_runtime": 458.0063, | |
| "eval_samples_per_second": 2.183, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.10879325025156746, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 1.273606324733284e-06, | |
| "loss": 0.0301, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 8.176284240242638e-07, | |
| "loss": 0.03, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "eval_f1_score": 0.6109148017275227, | |
| "eval_label_f1": 0.8370632116215156, | |
| "eval_loss": 0.04447592794895172, | |
| "eval_runtime": 458.4151, | |
| "eval_samples_per_second": 2.181, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 0.10871584487963465, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 4.6102107522336403e-07, | |
| "loss": 0.0307, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "learning_rate": 2.052496544188487e-07, | |
| "loss": 0.0296, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "eval_f1_score": 0.6117000392618768, | |
| "eval_label_f1": 0.8378484491558698, | |
| "eval_loss": 0.04443991929292679, | |
| "eval_runtime": 455.2323, | |
| "eval_samples_per_second": 2.197, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 0.1084062233919034, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 5.136518124159162e-08, | |
| "loss": 0.0288, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 0.0, | |
| "loss": 0.0297, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "eval_f1_score": 0.6098231827111984, | |
| "eval_label_f1": 0.8369351669941061, | |
| "eval_loss": 0.04443608224391937, | |
| "eval_runtime": 456.0761, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.10856103413576902, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 200, | |
| "total_flos": 1.9542812433710973e+20, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |