| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5097099750242112, | |
| "global_step": 20000, | |
| "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 | |
| } | |
| ], | |
| "max_steps": 500000, | |
| "num_train_epochs": 13, | |
| "total_flos": 6.38974277549e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |