| { | |
| "best_metric": 0.9040894380700206, | |
| "best_model_checkpoint": "trillsson3-ft-keyword-spotting-15/checkpoint-3192", | |
| "epoch": 9.999373825923607, | |
| "global_step": 7980, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.533834586466165e-05, | |
| "loss": 7.4799, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.293233082706766e-05, | |
| "loss": 4.7288, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00011052631578947366, | |
| "loss": 3.0455, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00014812030075187968, | |
| "loss": 2.3495, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00018571428571428572, | |
| "loss": 1.836, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00022330827067669172, | |
| "loss": 1.4179, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0002609022556390977, | |
| "loss": 1.1824, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7488967343336276, | |
| "eval_loss": 0.6477929949760437, | |
| "eval_runtime": 41.7521, | |
| "eval_samples_per_second": 162.818, | |
| "eval_steps_per_second": 2.563, | |
| "step": 798 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002984962406015037, | |
| "loss": 1.0261, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00029598997493734334, | |
| "loss": 0.9596, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.000291812865497076, | |
| "loss": 0.8732, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00028763575605680864, | |
| "loss": 0.8509, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00028345864661654134, | |
| "loss": 0.8162, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.000279281537176274, | |
| "loss": 0.7603, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00027510442773600664, | |
| "loss": 0.7638, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00027092731829573934, | |
| "loss": 0.7448, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8727566931450427, | |
| "eval_loss": 0.4273872375488281, | |
| "eval_runtime": 40.9071, | |
| "eval_samples_per_second": 166.181, | |
| "eval_steps_per_second": 2.616, | |
| "step": 1596 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.000266750208855472, | |
| "loss": 0.7268, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00026257309941520463, | |
| "loss": 0.7503, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.00025839598997493734, | |
| "loss": 0.719, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.00025421888053467, | |
| "loss": 0.7174, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.00025004177109440263, | |
| "loss": 0.6869, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.00024586466165413533, | |
| "loss": 0.7113, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.00024168755221386798, | |
| "loss": 0.7011, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00023751044277360066, | |
| "loss": 0.7089, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8949691085613416, | |
| "eval_loss": 0.3723289370536804, | |
| "eval_runtime": 42.3115, | |
| "eval_samples_per_second": 160.665, | |
| "eval_steps_per_second": 2.529, | |
| "step": 2394 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0002333333333333333, | |
| "loss": 0.6834, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.00022915622389306598, | |
| "loss": 0.6804, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 0.00022497911445279865, | |
| "loss": 0.6771, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.0002208020050125313, | |
| "loss": 0.684, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.00021662489557226398, | |
| "loss": 0.6636, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 0.00021244778613199665, | |
| "loss": 0.7003, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.0002082706766917293, | |
| "loss": 0.6594, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 0.00020409356725146197, | |
| "loss": 0.6781, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9040894380700206, | |
| "eval_loss": 0.3562973439693451, | |
| "eval_runtime": 42.1998, | |
| "eval_samples_per_second": 161.091, | |
| "eval_steps_per_second": 2.536, | |
| "step": 3192 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 0.00019991645781119465, | |
| "loss": 0.6921, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.0001957393483709273, | |
| "loss": 0.6615, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00019156223893065997, | |
| "loss": 0.6707, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 0.00018738512949039265, | |
| "loss": 0.6511, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.0001832080200501253, | |
| "loss": 0.6974, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.00017903091060985797, | |
| "loss": 0.6447, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.00017485380116959065, | |
| "loss": 0.643, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 0.0001706766917293233, | |
| "loss": 0.6386, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8986466607825832, | |
| "eval_loss": 0.3440994322299957, | |
| "eval_runtime": 41.4342, | |
| "eval_samples_per_second": 164.067, | |
| "eval_steps_per_second": 2.582, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.00016649958228905597, | |
| "loss": 0.6611, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.0001623642439431913, | |
| "loss": 0.6634, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 0.00015818713450292397, | |
| "loss": 0.6419, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 0.00015401002506265662, | |
| "loss": 0.6341, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 0.0001498329156223893, | |
| "loss": 0.6283, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 0.00014565580618212197, | |
| "loss": 0.6551, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.00014147869674185462, | |
| "loss": 0.6318, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.0001373015873015873, | |
| "loss": 0.6342, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8993821712268314, | |
| "eval_loss": 0.3380272388458252, | |
| "eval_runtime": 42.3543, | |
| "eval_samples_per_second": 160.503, | |
| "eval_steps_per_second": 2.526, | |
| "step": 4788 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 0.00013312447786131997, | |
| "loss": 0.6371, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 0.00012894736842105261, | |
| "loss": 0.6484, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 0.0001247702589807853, | |
| "loss": 0.6331, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 0.00012059314954051795, | |
| "loss": 0.6201, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 0.00011641604010025061, | |
| "loss": 0.6619, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 0.00011223893065998329, | |
| "loss": 0.6337, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 0.00010806182121971595, | |
| "loss": 0.6356, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.00010388471177944861, | |
| "loss": 0.6275, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8982053545160341, | |
| "eval_loss": 0.33755984902381897, | |
| "eval_runtime": 43.6608, | |
| "eval_samples_per_second": 155.7, | |
| "eval_steps_per_second": 2.451, | |
| "step": 5586 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 9.970760233918129e-05, | |
| "loss": 0.6359, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 9.553049289891395e-05, | |
| "loss": 0.6502, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 9.135338345864661e-05, | |
| "loss": 0.6138, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 8.717627401837928e-05, | |
| "loss": 0.6309, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 8.299916457811194e-05, | |
| "loss": 0.6386, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 7.88220551378446e-05, | |
| "loss": 0.6149, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 7.464494569757727e-05, | |
| "loss": 0.6218, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 7.050960735171261e-05, | |
| "loss": 0.6349, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9014416004707266, | |
| "eval_loss": 0.3333294689655304, | |
| "eval_runtime": 41.9257, | |
| "eval_samples_per_second": 162.144, | |
| "eval_steps_per_second": 2.552, | |
| "step": 6384 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 6.633249791144527e-05, | |
| "loss": 0.6115, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 6.215538847117793e-05, | |
| "loss": 0.6463, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "learning_rate": 5.79782790309106e-05, | |
| "loss": 0.6183, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 5.380116959064327e-05, | |
| "loss": 0.6475, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 4.962406015037593e-05, | |
| "loss": 0.6437, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 4.54469507101086e-05, | |
| "loss": 0.6265, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 4.1269841269841266e-05, | |
| "loss": 0.6326, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 3.709273182957393e-05, | |
| "loss": 0.6261, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9024713150926743, | |
| "eval_loss": 0.32949280738830566, | |
| "eval_runtime": 43.6022, | |
| "eval_samples_per_second": 155.91, | |
| "eval_steps_per_second": 2.454, | |
| "step": 7182 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.2915622389306596e-05, | |
| "loss": 0.6306, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 2.873851294903926e-05, | |
| "loss": 0.6362, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 2.4561403508771925e-05, | |
| "loss": 0.6318, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 2.0384294068504594e-05, | |
| "loss": 0.6506, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 1.620718462823726e-05, | |
| "loss": 0.6202, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 1.2030075187969923e-05, | |
| "loss": 0.6217, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 7.852965747702588e-06, | |
| "loss": 0.6307, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 3.6758563074352545e-06, | |
| "loss": 0.6188, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9024713150926743, | |
| "eval_loss": 0.33220550417900085, | |
| "eval_runtime": 43.7783, | |
| "eval_samples_per_second": 155.282, | |
| "eval_steps_per_second": 2.444, | |
| "step": 7980 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 7980, | |
| "total_flos": 0.0, | |
| "train_loss": 0.8915880279732229, | |
| "train_runtime": 4549.8269, | |
| "train_samples_per_second": 112.299, | |
| "train_steps_per_second": 1.754 | |
| } | |
| ], | |
| "max_steps": 7980, | |
| "num_train_epochs": 10, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |