| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 11250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.4299999999999998e-06, | |
| "loss": 12.3928, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 6.93e-06, | |
| "loss": 7.13, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.0429999999999998e-05, | |
| "loss": 4.5055, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.3929999999999999e-05, | |
| "loss": 3.8494, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.7429999999999997e-05, | |
| "loss": 3.4447, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.0929999999999998e-05, | |
| "loss": 3.2021, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 2.4429999999999995e-05, | |
| "loss": 3.0761, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.793e-05, | |
| "loss": 3.0165, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.1429999999999996e-05, | |
| "loss": 2.9623, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 3.493e-05, | |
| "loss": 2.9094, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 3.843e-05, | |
| "loss": 2.8831, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 4.192999999999999e-05, | |
| "loss": 2.8667, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 4.543e-05, | |
| "loss": 2.7722, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 4.8929999999999994e-05, | |
| "loss": 2.2496, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 5.243e-05, | |
| "loss": 1.7213, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 5.593e-05, | |
| "loss": 1.5195, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 5.942999999999999e-05, | |
| "loss": 1.3847, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.293e-05, | |
| "loss": 1.3296, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 6.642999999999999e-05, | |
| "loss": 1.2533, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 6.992999999999999e-05, | |
| "loss": 1.1973, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "eval_loss": 0.44813069701194763, | |
| "eval_runtime": 151.5294, | |
| "eval_samples_per_second": 20.933, | |
| "eval_steps_per_second": 20.933, | |
| "eval_wer": 0.4849491021888243, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 6.925837837837837e-05, | |
| "loss": 1.08, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 6.850162162162162e-05, | |
| "loss": 0.9484, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 6.774486486486486e-05, | |
| "loss": 0.8818, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "learning_rate": 6.698810810810811e-05, | |
| "loss": 0.8043, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 6.623135135135134e-05, | |
| "loss": 0.7752, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 6.547459459459459e-05, | |
| "loss": 0.7497, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 6.471783783783783e-05, | |
| "loss": 0.7486, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 6.396108108108108e-05, | |
| "loss": 0.7016, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "learning_rate": 6.320432432432433e-05, | |
| "loss": 0.6927, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 6.244756756756756e-05, | |
| "loss": 0.6609, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 13.78, | |
| "learning_rate": 6.169081081081081e-05, | |
| "loss": 0.6633, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 14.22, | |
| "learning_rate": 6.093405405405405e-05, | |
| "loss": 0.6781, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 14.67, | |
| "learning_rate": 6.017729729729729e-05, | |
| "loss": 0.6486, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 5.942054054054054e-05, | |
| "loss": 0.6217, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 5.866378378378378e-05, | |
| "loss": 0.6348, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 5.7907027027027026e-05, | |
| "loss": 0.6555, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "learning_rate": 5.715027027027027e-05, | |
| "loss": 0.6179, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 5.639351351351351e-05, | |
| "loss": 0.6116, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 17.33, | |
| "learning_rate": 5.5636756756756754e-05, | |
| "loss": 0.586, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 5.4879999999999996e-05, | |
| "loss": 0.6005, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "eval_loss": 0.1420038342475891, | |
| "eval_runtime": 156.5701, | |
| "eval_samples_per_second": 20.259, | |
| "eval_steps_per_second": 20.259, | |
| "eval_wer": 0.17772729258595832, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 18.22, | |
| "learning_rate": 5.412324324324324e-05, | |
| "loss": 0.6199, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "learning_rate": 5.336648648648648e-05, | |
| "loss": 0.6017, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 19.11, | |
| "learning_rate": 5.2609729729729724e-05, | |
| "loss": 0.5722, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 19.56, | |
| "learning_rate": 5.1852972972972974e-05, | |
| "loss": 0.5755, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.1096216216216216e-05, | |
| "loss": 0.6083, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "learning_rate": 5.033945945945946e-05, | |
| "loss": 0.5677, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 20.89, | |
| "learning_rate": 4.9590270270270266e-05, | |
| "loss": 0.5652, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 21.33, | |
| "learning_rate": 4.8833513513513516e-05, | |
| "loss": 0.5509, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 21.78, | |
| "learning_rate": 4.807675675675676e-05, | |
| "loss": 0.5526, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 4.732e-05, | |
| "loss": 0.5589, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 22.67, | |
| "learning_rate": 4.6563243243243244e-05, | |
| "loss": 0.5394, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "learning_rate": 4.5806486486486486e-05, | |
| "loss": 0.5329, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 23.56, | |
| "learning_rate": 4.504972972972973e-05, | |
| "loss": 0.5353, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 4.429297297297297e-05, | |
| "loss": 0.5643, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "learning_rate": 4.3536216216216214e-05, | |
| "loss": 0.537, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "learning_rate": 4.277945945945946e-05, | |
| "loss": 0.5502, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 25.33, | |
| "learning_rate": 4.20227027027027e-05, | |
| "loss": 0.5126, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 25.78, | |
| "learning_rate": 4.126594594594594e-05, | |
| "loss": 0.5315, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 26.22, | |
| "learning_rate": 4.050918918918919e-05, | |
| "loss": 0.5424, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 3.9752432432432434e-05, | |
| "loss": 0.5248, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "eval_loss": 0.13026614487171173, | |
| "eval_runtime": 153.4664, | |
| "eval_samples_per_second": 20.669, | |
| "eval_steps_per_second": 20.669, | |
| "eval_wer": 0.16505745117742146, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 27.11, | |
| "learning_rate": 3.8995675675675676e-05, | |
| "loss": 0.5111, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 27.56, | |
| "learning_rate": 3.823891891891892e-05, | |
| "loss": 0.5226, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 3.748216216216216e-05, | |
| "loss": 0.5335, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 28.44, | |
| "learning_rate": 3.6725405405405404e-05, | |
| "loss": 0.5031, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "learning_rate": 3.596864864864865e-05, | |
| "loss": 0.5219, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 29.33, | |
| "learning_rate": 3.521189189189189e-05, | |
| "loss": 0.4853, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 29.78, | |
| "learning_rate": 3.445513513513513e-05, | |
| "loss": 0.5062, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 30.22, | |
| "learning_rate": 3.370594594594594e-05, | |
| "loss": 0.5395, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 30.67, | |
| "learning_rate": 3.294918918918919e-05, | |
| "loss": 0.4876, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 31.11, | |
| "learning_rate": 3.219243243243243e-05, | |
| "loss": 0.4981, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 31.56, | |
| "learning_rate": 3.1435675675675674e-05, | |
| "loss": 0.5011, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 3.067891891891892e-05, | |
| "loss": 0.511, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 32.44, | |
| "learning_rate": 2.992216216216216e-05, | |
| "loss": 0.4935, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 32.89, | |
| "learning_rate": 2.9165405405405402e-05, | |
| "loss": 0.4951, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 2.8408648648648645e-05, | |
| "loss": 0.4655, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 33.78, | |
| "learning_rate": 2.765189189189189e-05, | |
| "loss": 0.4926, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 34.22, | |
| "learning_rate": 2.6895135135135133e-05, | |
| "loss": 0.5083, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "learning_rate": 2.6138378378378376e-05, | |
| "loss": 0.4849, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 35.11, | |
| "learning_rate": 2.538162162162162e-05, | |
| "loss": 0.4673, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "learning_rate": 2.462486486486486e-05, | |
| "loss": 0.4871, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "eval_loss": 0.12074683606624603, | |
| "eval_runtime": 154.3437, | |
| "eval_samples_per_second": 20.552, | |
| "eval_steps_per_second": 20.552, | |
| "eval_wer": 0.1523439206605793, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 2.3875675675675676e-05, | |
| "loss": 0.4911, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 36.44, | |
| "learning_rate": 2.3118918918918918e-05, | |
| "loss": 0.4724, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 36.89, | |
| "learning_rate": 2.236216216216216e-05, | |
| "loss": 0.4784, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 37.33, | |
| "learning_rate": 2.1605405405405403e-05, | |
| "loss": 0.466, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 37.78, | |
| "learning_rate": 2.0848648648648646e-05, | |
| "loss": 0.4761, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 38.22, | |
| "learning_rate": 2.0091891891891892e-05, | |
| "loss": 0.4772, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 38.67, | |
| "learning_rate": 1.9335135135135135e-05, | |
| "loss": 0.4524, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 39.11, | |
| "learning_rate": 1.8578378378378377e-05, | |
| "loss": 0.4436, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 39.56, | |
| "learning_rate": 1.782162162162162e-05, | |
| "loss": 0.4673, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 1.7064864864864862e-05, | |
| "loss": 0.4848, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 40.44, | |
| "learning_rate": 1.630810810810811e-05, | |
| "loss": 0.461, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 40.89, | |
| "learning_rate": 1.555135135135135e-05, | |
| "loss": 0.465, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 41.33, | |
| "learning_rate": 1.4794594594594594e-05, | |
| "loss": 0.4398, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 41.78, | |
| "learning_rate": 1.4045405405405405e-05, | |
| "loss": 0.4552, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 42.22, | |
| "learning_rate": 1.3288648648648647e-05, | |
| "loss": 0.47, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 42.67, | |
| "learning_rate": 1.253189189189189e-05, | |
| "loss": 0.4599, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 43.11, | |
| "learning_rate": 1.1775135135135134e-05, | |
| "loss": 0.4273, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 43.56, | |
| "learning_rate": 1.1018378378378377e-05, | |
| "loss": 0.4533, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 1.0261621621621621e-05, | |
| "loss": 0.4573, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "learning_rate": 9.504864864864864e-06, | |
| "loss": 0.4428, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "eval_loss": 0.11431078612804413, | |
| "eval_runtime": 152.8495, | |
| "eval_samples_per_second": 20.752, | |
| "eval_steps_per_second": 20.752, | |
| "eval_wer": 0.14247018218358162, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 44.89, | |
| "learning_rate": 8.748108108108106e-06, | |
| "loss": 0.4431, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 45.33, | |
| "learning_rate": 7.99135135135135e-06, | |
| "loss": 0.4124, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 45.78, | |
| "learning_rate": 7.234594594594593e-06, | |
| "loss": 0.4437, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 46.22, | |
| "learning_rate": 6.4778378378378375e-06, | |
| "loss": 0.4694, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 46.67, | |
| "learning_rate": 5.721081081081081e-06, | |
| "loss": 0.4408, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 47.11, | |
| "learning_rate": 4.9643243243243245e-06, | |
| "loss": 0.428, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 47.56, | |
| "learning_rate": 4.207567567567567e-06, | |
| "loss": 0.4418, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 3.4508108108108105e-06, | |
| "loss": 0.4527, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 48.44, | |
| "learning_rate": 2.6940540540540536e-06, | |
| "loss": 0.448, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 48.89, | |
| "learning_rate": 1.937297297297297e-06, | |
| "loss": 0.4399, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 49.33, | |
| "learning_rate": 1.1805405405405403e-06, | |
| "loss": 0.4111, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 49.78, | |
| "learning_rate": 4.237837837837838e-07, | |
| "loss": 0.4214, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 11250, | |
| "total_flos": 4.148416605366081e+19, | |
| "train_loss": 1.0303706246270075, | |
| "train_runtime": 15356.7914, | |
| "train_samples_per_second": 23.358, | |
| "train_steps_per_second": 0.733 | |
| } | |
| ], | |
| "max_steps": 11250, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.148416605366081e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |