| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7839849125847391, | |
| "global_step": 70000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5.999999999999999e-06, | |
| "loss": 0.8907, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.1999999999999999e-05, | |
| "loss": 0.7216, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.9141247272491455, | |
| "eval_runtime": 1.2377, | |
| "eval_samples_per_second": 807.918, | |
| "eval_steps_per_second": 12.927, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.7999999999999997e-05, | |
| "loss": 0.7122, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.3999999999999997e-05, | |
| "loss": 0.7114, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.9156034588813782, | |
| "eval_runtime": 1.1631, | |
| "eval_samples_per_second": 859.777, | |
| "eval_steps_per_second": 13.756, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 0.7111, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3.5999999999999994e-05, | |
| "loss": 0.711, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.915696918964386, | |
| "eval_runtime": 1.149, | |
| "eval_samples_per_second": 870.294, | |
| "eval_steps_per_second": 13.925, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.7107, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.7999999999999994e-05, | |
| "loss": 0.7102, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.8820520043373108, | |
| "eval_runtime": 1.2331, | |
| "eval_samples_per_second": 810.943, | |
| "eval_steps_per_second": 12.975, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5.399999999999999e-05, | |
| "loss": 0.7096, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 0.709, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.8773962259292603, | |
| "eval_runtime": 1.22, | |
| "eval_samples_per_second": 819.664, | |
| "eval_steps_per_second": 13.115, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 6.599999999999999e-05, | |
| "loss": 0.7087, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 7.199999999999999e-05, | |
| "loss": 0.7083, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.8944988250732422, | |
| "eval_runtime": 1.1878, | |
| "eval_samples_per_second": 841.893, | |
| "eval_steps_per_second": 13.47, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 7.8e-05, | |
| "loss": 0.6868, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.6516, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.8944807648658752, | |
| "eval_runtime": 1.2213, | |
| "eval_samples_per_second": 818.801, | |
| "eval_steps_per_second": 13.101, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 0.6267, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.599999999999999e-05, | |
| "loss": 0.6042, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.8855597972869873, | |
| "eval_runtime": 1.1662, | |
| "eval_samples_per_second": 857.462, | |
| "eval_steps_per_second": 13.719, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000102, | |
| "loss": 0.5883, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00010799999999999998, | |
| "loss": 0.5732, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.8907042741775513, | |
| "eval_runtime": 1.2023, | |
| "eval_samples_per_second": 831.767, | |
| "eval_steps_per_second": 13.308, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00011399999999999999, | |
| "loss": 0.559, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.5506, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.8921875953674316, | |
| "eval_runtime": 1.197, | |
| "eval_samples_per_second": 835.455, | |
| "eval_steps_per_second": 13.367, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00012599999999999997, | |
| "loss": 0.5444, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00013199999999999998, | |
| "loss": 0.5385, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.8903993964195251, | |
| "eval_runtime": 1.1996, | |
| "eval_samples_per_second": 833.63, | |
| "eval_steps_per_second": 13.338, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.000138, | |
| "loss": 0.532, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00014399999999999998, | |
| "loss": 0.5276, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.8864779472351074, | |
| "eval_runtime": 1.1846, | |
| "eval_samples_per_second": 844.173, | |
| "eval_steps_per_second": 13.507, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00015, | |
| "loss": 0.5218, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.000156, | |
| "loss": 0.517, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.88374263048172, | |
| "eval_runtime": 1.1943, | |
| "eval_samples_per_second": 837.343, | |
| "eval_steps_per_second": 13.397, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.000162, | |
| "loss": 0.5124, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.000168, | |
| "loss": 0.5077, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 0.8863275647163391, | |
| "eval_runtime": 1.1844, | |
| "eval_samples_per_second": 844.316, | |
| "eval_steps_per_second": 13.509, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00017399999999999997, | |
| "loss": 0.5027, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.498, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.8769687414169312, | |
| "eval_runtime": 1.1484, | |
| "eval_samples_per_second": 870.806, | |
| "eval_steps_per_second": 13.933, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000186, | |
| "loss": 0.4938, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00019199999999999998, | |
| "loss": 0.4897, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 0.8793530464172363, | |
| "eval_runtime": 1.2247, | |
| "eval_samples_per_second": 816.516, | |
| "eval_steps_per_second": 13.064, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.000198, | |
| "loss": 0.4849, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.000204, | |
| "loss": 0.4791, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 0.879580557346344, | |
| "eval_runtime": 1.1672, | |
| "eval_samples_per_second": 856.75, | |
| "eval_steps_per_second": 13.708, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.4744, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00021599999999999996, | |
| "loss": 0.4698, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 0.8753651976585388, | |
| "eval_runtime": 1.1516, | |
| "eval_samples_per_second": 868.392, | |
| "eval_steps_per_second": 13.894, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00022199999999999998, | |
| "loss": 0.4644, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00022799999999999999, | |
| "loss": 0.4592, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.8825145363807678, | |
| "eval_runtime": 1.2668, | |
| "eval_samples_per_second": 789.404, | |
| "eval_steps_per_second": 12.63, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000234, | |
| "loss": 0.4542, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.4489, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.878689169883728, | |
| "eval_runtime": 1.1589, | |
| "eval_samples_per_second": 862.885, | |
| "eval_steps_per_second": 13.806, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00024599999999999996, | |
| "loss": 0.4437, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00025199999999999995, | |
| "loss": 0.439, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 0.8742354512214661, | |
| "eval_runtime": 1.1831, | |
| "eval_samples_per_second": 845.245, | |
| "eval_steps_per_second": 13.524, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.000258, | |
| "loss": 0.4339, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00026399999999999997, | |
| "loss": 0.4292, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.8848758935928345, | |
| "eval_runtime": 1.1886, | |
| "eval_samples_per_second": 841.313, | |
| "eval_steps_per_second": 13.461, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00027, | |
| "loss": 0.4251, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.000276, | |
| "loss": 0.4212, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": 0.8841528296470642, | |
| "eval_runtime": 1.2211, | |
| "eval_samples_per_second": 818.95, | |
| "eval_steps_per_second": 13.103, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00028199999999999997, | |
| "loss": 0.4175, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00028799999999999995, | |
| "loss": 0.4142, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.8811973929405212, | |
| "eval_runtime": 1.275, | |
| "eval_samples_per_second": 784.285, | |
| "eval_steps_per_second": 12.549, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.000294, | |
| "loss": 0.4109, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4076, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.8659318089485168, | |
| "eval_runtime": 1.2881, | |
| "eval_samples_per_second": 776.338, | |
| "eval_steps_per_second": 12.421, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00029999920715161553, | |
| "loss": 0.4045, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002999968286151326, | |
| "loss": 0.4017, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 0.874371349811554, | |
| "eval_runtime": 1.2282, | |
| "eval_samples_per_second": 814.215, | |
| "eval_steps_per_second": 13.027, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0002999928644165624, | |
| "loss": 0.3987, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0002999873145992569, | |
| "loss": 0.3958, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 0.8821514844894409, | |
| "eval_runtime": 1.1414, | |
| "eval_samples_per_second": 876.15, | |
| "eval_steps_per_second": 14.018, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.000299980179223908, | |
| "loss": 0.3933, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002999714583685469, | |
| "loss": 0.3907, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.8762433528900146, | |
| "eval_runtime": 1.2113, | |
| "eval_samples_per_second": 825.526, | |
| "eval_steps_per_second": 13.208, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00029996115212854366, | |
| "loss": 0.3886, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00029994926061660554, | |
| "loss": 0.3863, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_loss": 0.8757530450820923, | |
| "eval_runtime": 1.1818, | |
| "eval_samples_per_second": 846.134, | |
| "eval_steps_per_second": 13.538, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0002999357839627762, | |
| "loss": 0.3841, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00029992072231443425, | |
| "loss": 0.382, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.8755055069923401, | |
| "eval_runtime": 1.2044, | |
| "eval_samples_per_second": 830.296, | |
| "eval_steps_per_second": 13.285, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002999040758362914, | |
| "loss": 0.3804, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00029988584471039094, | |
| "loss": 0.378, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 0.8780514001846313, | |
| "eval_runtime": 1.1913, | |
| "eval_samples_per_second": 839.425, | |
| "eval_steps_per_second": 13.431, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002998660291361054, | |
| "loss": 0.3763, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0002998446293301349, | |
| "loss": 0.3748, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 0.8814873099327087, | |
| "eval_runtime": 1.3032, | |
| "eval_samples_per_second": 767.316, | |
| "eval_steps_per_second": 12.277, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.0002998216455265042, | |
| "loss": 0.3735, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00029979707797656046, | |
| "loss": 0.3716, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.868879497051239, | |
| "eval_runtime": 1.2062, | |
| "eval_samples_per_second": 829.022, | |
| "eval_steps_per_second": 13.264, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00029977092694897053, | |
| "loss": 0.3702, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002997431927297178, | |
| "loss": 0.3689, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 0.8758594989776611, | |
| "eval_runtime": 1.2131, | |
| "eval_samples_per_second": 824.364, | |
| "eval_steps_per_second": 13.19, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00029971387562209936, | |
| "loss": 0.3677, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00029968297594672226, | |
| "loss": 0.3665, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.8690294623374939, | |
| "eval_runtime": 1.1601, | |
| "eval_samples_per_second": 861.986, | |
| "eval_steps_per_second": 13.792, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0002996504940415005, | |
| "loss": 0.365, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00029961643026165096, | |
| "loss": 0.364, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 0.8695961236953735, | |
| "eval_runtime": 1.2184, | |
| "eval_samples_per_second": 820.732, | |
| "eval_steps_per_second": 13.132, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00029958078497968973, | |
| "loss": 0.3627, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002995435585854278, | |
| "loss": 0.3614, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 0.8684411644935608, | |
| "eval_runtime": 1.2551, | |
| "eval_samples_per_second": 796.719, | |
| "eval_steps_per_second": 12.747, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0002995047514859671, | |
| "loss": 0.3603, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0002994643641056959, | |
| "loss": 0.3592, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.8597822189331055, | |
| "eval_runtime": 1.198, | |
| "eval_samples_per_second": 834.721, | |
| "eval_steps_per_second": 13.356, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.000299422396886284, | |
| "loss": 0.358, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.0002993788502866783, | |
| "loss": 0.3571, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 0.8572097420692444, | |
| "eval_runtime": 1.21, | |
| "eval_samples_per_second": 826.463, | |
| "eval_steps_per_second": 13.223, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00029933372478309746, | |
| "loss": 0.3562, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00029928702086902664, | |
| "loss": 0.3555, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.8637193441390991, | |
| "eval_runtime": 1.4051, | |
| "eval_samples_per_second": 711.674, | |
| "eval_steps_per_second": 11.387, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00029923873905521244, | |
| "loss": 0.3545, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.000299188879869657, | |
| "loss": 0.3535, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_loss": 0.8638287782669067, | |
| "eval_runtime": 1.2182, | |
| "eval_samples_per_second": 820.876, | |
| "eval_steps_per_second": 13.134, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00029913744385761244, | |
| "loss": 0.3524, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00029908443158157465, | |
| "loss": 0.3518, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.8664878606796265, | |
| "eval_runtime": 1.2547, | |
| "eval_samples_per_second": 796.987, | |
| "eval_steps_per_second": 12.752, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0002990298436212775, | |
| "loss": 0.3511, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0002989736805736861, | |
| "loss": 0.3502, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.8559480905532837, | |
| "eval_runtime": 1.2272, | |
| "eval_samples_per_second": 814.861, | |
| "eval_steps_per_second": 13.038, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00029891594305299065, | |
| "loss": 0.3494, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00029885663169059926, | |
| "loss": 0.3488, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.859957754611969, | |
| "eval_runtime": 1.2944, | |
| "eval_samples_per_second": 772.587, | |
| "eval_steps_per_second": 12.361, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0002987957471351316, | |
| "loss": 0.3478, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00029873329005241137, | |
| "loss": 0.3469, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 0.852756917476654, | |
| "eval_runtime": 1.2478, | |
| "eval_samples_per_second": 801.439, | |
| "eval_steps_per_second": 12.823, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00029866926112545925, | |
| "loss": 0.3464, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00029860366105448534, | |
| "loss": 0.3459, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 0.8597527146339417, | |
| "eval_runtime": 1.1814, | |
| "eval_samples_per_second": 846.435, | |
| "eval_steps_per_second": 13.543, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00029853649055688143, | |
| "loss": 0.3451, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00029846775036721337, | |
| "loss": 0.3444, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.860701322555542, | |
| "eval_runtime": 1.2083, | |
| "eval_samples_per_second": 827.633, | |
| "eval_steps_per_second": 13.242, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.0002983974412372129, | |
| "loss": 0.3438, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00029832556393576934, | |
| "loss": 0.3428, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_loss": 0.865045964717865, | |
| "eval_runtime": 1.2493, | |
| "eval_samples_per_second": 800.445, | |
| "eval_steps_per_second": 12.807, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002982521192489214, | |
| "loss": 0.3425, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0002981771079798483, | |
| "loss": 0.342, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 0.8640099167823792, | |
| "eval_runtime": 1.2369, | |
| "eval_samples_per_second": 808.489, | |
| "eval_steps_per_second": 12.936, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00029810053094886136, | |
| "loss": 0.3417, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00029802238899339473, | |
| "loss": 0.3408, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.8549481630325317, | |
| "eval_runtime": 1.3108, | |
| "eval_samples_per_second": 762.871, | |
| "eval_steps_per_second": 12.206, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0002979426829679962, | |
| "loss": 0.3403, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0002978614137443183, | |
| "loss": 0.3398, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.8630042672157288, | |
| "eval_runtime": 1.1917, | |
| "eval_samples_per_second": 839.116, | |
| "eval_steps_per_second": 13.426, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.000297778582211108, | |
| "loss": 0.3391, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00029769418927419786, | |
| "loss": 0.3387, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 0.8540878891944885, | |
| "eval_runtime": 1.2825, | |
| "eval_samples_per_second": 779.707, | |
| "eval_steps_per_second": 12.475, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0002976082358564954, | |
| "loss": 0.338, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00029752072289797353, | |
| "loss": 0.3373, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_loss": 0.8587937355041504, | |
| "eval_runtime": 1.3038, | |
| "eval_samples_per_second": 766.964, | |
| "eval_steps_per_second": 12.271, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00029743165135565986, | |
| "loss": 0.3373, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00029734102220362654, | |
| "loss": 0.3368, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 0.8638780117034912, | |
| "eval_runtime": 1.2508, | |
| "eval_samples_per_second": 799.458, | |
| "eval_steps_per_second": 12.791, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00029724883643297937, | |
| "loss": 0.3362, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0002971550950518473, | |
| "loss": 0.3357, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 0.8546432256698608, | |
| "eval_runtime": 1.2572, | |
| "eval_samples_per_second": 795.415, | |
| "eval_steps_per_second": 12.727, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.000297059799085371, | |
| "loss": 0.3351, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00029696294957569196, | |
| "loss": 0.335, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 0.8535034656524658, | |
| "eval_runtime": 1.2455, | |
| "eval_samples_per_second": 802.916, | |
| "eval_steps_per_second": 12.847, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00029686454758194076, | |
| "loss": 0.3343, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00029676459418022594, | |
| "loss": 0.334, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.8510599136352539, | |
| "eval_runtime": 1.2497, | |
| "eval_samples_per_second": 800.205, | |
| "eval_steps_per_second": 12.803, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002966630904636219, | |
| "loss": 0.3335, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.0002965600375421569, | |
| "loss": 0.333, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.8525308966636658, | |
| "eval_runtime": 1.285, | |
| "eval_samples_per_second": 778.223, | |
| "eval_steps_per_second": 12.452, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0002964554365428013, | |
| "loss": 0.3328, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00029634928860945486, | |
| "loss": 0.3322, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 0.8536081910133362, | |
| "eval_runtime": 1.2803, | |
| "eval_samples_per_second": 781.08, | |
| "eval_steps_per_second": 12.497, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.0002962415949029343, | |
| "loss": 0.3319, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00029613235660096084, | |
| "loss": 0.3314, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 0.8421887159347534, | |
| "eval_runtime": 1.3286, | |
| "eval_samples_per_second": 752.666, | |
| "eval_steps_per_second": 12.043, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00029602157489814693, | |
| "loss": 0.3311, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.00029590925100598365, | |
| "loss": 0.3307, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 0.8626954555511475, | |
| "eval_runtime": 1.268, | |
| "eval_samples_per_second": 788.647, | |
| "eval_steps_per_second": 12.618, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.000295795386152827, | |
| "loss": 0.3302, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.0002956799815838848, | |
| "loss": 0.3298, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_loss": 0.8434558510780334, | |
| "eval_runtime": 1.2564, | |
| "eval_samples_per_second": 795.92, | |
| "eval_steps_per_second": 12.735, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0002955630385612029, | |
| "loss": 0.3294, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0002954445583636515, | |
| "loss": 0.3292, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.8569065928459167, | |
| "eval_runtime": 1.2969, | |
| "eval_samples_per_second": 771.061, | |
| "eval_steps_per_second": 12.337, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00029532454228691103, | |
| "loss": 0.3288, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.0002952029916434581, | |
| "loss": 0.3287, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_loss": 0.8517204523086548, | |
| "eval_runtime": 1.2275, | |
| "eval_samples_per_second": 814.648, | |
| "eval_steps_per_second": 13.034, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00029507990776255107, | |
| "loss": 0.3282, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00029495529199021555, | |
| "loss": 0.3278, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_loss": 0.8487727046012878, | |
| "eval_runtime": 1.2356, | |
| "eval_samples_per_second": 809.327, | |
| "eval_steps_per_second": 12.949, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0002948291456892296, | |
| "loss": 0.3273, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00029470147023910907, | |
| "loss": 0.3269, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 0.8469608426094055, | |
| "eval_runtime": 1.2425, | |
| "eval_samples_per_second": 804.832, | |
| "eval_steps_per_second": 12.877, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0002945722670360921, | |
| "loss": 0.3265, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0002944415374931243, | |
| "loss": 0.3262, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_loss": 0.8486921191215515, | |
| "eval_runtime": 1.2218, | |
| "eval_samples_per_second": 818.455, | |
| "eval_steps_per_second": 13.095, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00029430928303984295, | |
| "loss": 0.326, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002941755051225616, | |
| "loss": 0.3257, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.8429991602897644, | |
| "eval_runtime": 1.2262, | |
| "eval_samples_per_second": 815.556, | |
| "eval_steps_per_second": 13.049, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00029404020520425417, | |
| "loss": 0.3253, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.0002939033847645388, | |
| "loss": 0.325, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.8381994366645813, | |
| "eval_runtime": 1.2665, | |
| "eval_samples_per_second": 789.582, | |
| "eval_steps_per_second": 12.633, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00029376504529966195, | |
| "loss": 0.3246, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00029362518832248184, | |
| "loss": 0.3248, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 0.848027229309082, | |
| "eval_runtime": 1.243, | |
| "eval_samples_per_second": 804.488, | |
| "eval_steps_per_second": 12.872, | |
| "step": 70000 | |
| } | |
| ], | |
| "max_steps": 500000, | |
| "num_train_epochs": 13, | |
| "total_flos": 2.236396492508455e+21, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |