| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.722424669879707, | |
| "eval_steps": 10000, | |
| "global_step": 380000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.83937751004016e-06, | |
| "loss": 1.4799, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.0304539203643799, | |
| "eval_runtime": 69.6877, | |
| "eval_samples_per_second": 1434.973, | |
| "eval_steps_per_second": 14.952, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.638654618473896e-06, | |
| "loss": 0.9736, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.9039866328239441, | |
| "eval_runtime": 65.2654, | |
| "eval_samples_per_second": 1532.205, | |
| "eval_steps_per_second": 15.966, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.437951807228917e-06, | |
| "loss": 0.8922, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.8560834527015686, | |
| "eval_runtime": 64.4642, | |
| "eval_samples_per_second": 1551.248, | |
| "eval_steps_per_second": 16.164, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.237228915662652e-06, | |
| "loss": 0.8531, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.8226799368858337, | |
| "eval_runtime": 64.3868, | |
| "eval_samples_per_second": 1553.114, | |
| "eval_steps_per_second": 16.183, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 9.036526104417672e-06, | |
| "loss": 0.829, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 0.8042359948158264, | |
| "eval_runtime": 64.3396, | |
| "eval_samples_per_second": 1554.254, | |
| "eval_steps_per_second": 16.195, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.835823293172691e-06, | |
| "loss": 0.811, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": 0.7950388789176941, | |
| "eval_runtime": 64.5291, | |
| "eval_samples_per_second": 1549.687, | |
| "eval_steps_per_second": 16.148, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 8.635120481927711e-06, | |
| "loss": 0.7976, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 0.7777369022369385, | |
| "eval_runtime": 64.862, | |
| "eval_samples_per_second": 1541.734, | |
| "eval_steps_per_second": 16.065, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 8.434417670682732e-06, | |
| "loss": 0.7872, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 0.7702258825302124, | |
| "eval_runtime": 64.4619, | |
| "eval_samples_per_second": 1551.303, | |
| "eval_steps_per_second": 16.165, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 8.233714859437752e-06, | |
| "loss": 0.778, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 0.760698139667511, | |
| "eval_runtime": 64.4008, | |
| "eval_samples_per_second": 1552.775, | |
| "eval_steps_per_second": 16.18, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.033012048192772e-06, | |
| "loss": 0.7695, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 0.7517786622047424, | |
| "eval_runtime": 64.56, | |
| "eval_samples_per_second": 1548.946, | |
| "eval_steps_per_second": 16.14, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 7.832309236947791e-06, | |
| "loss": 0.7638, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 0.7468145489692688, | |
| "eval_runtime": 64.9838, | |
| "eval_samples_per_second": 1538.846, | |
| "eval_steps_per_second": 16.035, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 7.631606425702813e-06, | |
| "loss": 0.7573, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 0.739628255367279, | |
| "eval_runtime": 64.2595, | |
| "eval_samples_per_second": 1556.191, | |
| "eval_steps_per_second": 16.216, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 7.4309036144578315e-06, | |
| "loss": 0.7512, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.7346537709236145, | |
| "eval_runtime": 64.911, | |
| "eval_samples_per_second": 1540.57, | |
| "eval_steps_per_second": 16.053, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 7.2301807228915665e-06, | |
| "loss": 0.7468, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.7300976514816284, | |
| "eval_runtime": 64.8445, | |
| "eval_samples_per_second": 1542.15, | |
| "eval_steps_per_second": 16.069, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 7.029477911646587e-06, | |
| "loss": 0.7418, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 0.7252103686332703, | |
| "eval_runtime": 64.5503, | |
| "eval_samples_per_second": 1549.179, | |
| "eval_steps_per_second": 16.142, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 6.828775100401607e-06, | |
| "loss": 0.7379, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 0.7203709483146667, | |
| "eval_runtime": 64.4763, | |
| "eval_samples_per_second": 1550.958, | |
| "eval_steps_per_second": 16.161, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 6.628072289156627e-06, | |
| "loss": 0.7333, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 0.7167079448699951, | |
| "eval_runtime": 64.7852, | |
| "eval_samples_per_second": 1543.562, | |
| "eval_steps_per_second": 16.084, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 6.427369477911647e-06, | |
| "loss": 0.7298, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.7129059433937073, | |
| "eval_runtime": 64.609, | |
| "eval_samples_per_second": 1547.772, | |
| "eval_steps_per_second": 16.128, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 6.2266666666666675e-06, | |
| "loss": 0.7258, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.7089965343475342, | |
| "eval_runtime": 64.6825, | |
| "eval_samples_per_second": 1546.013, | |
| "eval_steps_per_second": 16.109, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 6.025963855421687e-06, | |
| "loss": 0.7229, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_loss": 0.7047411799430847, | |
| "eval_runtime": 64.3501, | |
| "eval_samples_per_second": 1553.999, | |
| "eval_steps_per_second": 16.193, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 5.825261044176708e-06, | |
| "loss": 0.7194, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 0.7026636004447937, | |
| "eval_runtime": 64.4785, | |
| "eval_samples_per_second": 1550.905, | |
| "eval_steps_per_second": 16.16, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.624558232931727e-06, | |
| "loss": 0.7171, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "eval_loss": 0.702177107334137, | |
| "eval_runtime": 64.5209, | |
| "eval_samples_per_second": 1549.886, | |
| "eval_steps_per_second": 16.15, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 5.423855421686748e-06, | |
| "loss": 0.7138, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 0.6978575587272644, | |
| "eval_runtime": 64.573, | |
| "eval_samples_per_second": 1548.634, | |
| "eval_steps_per_second": 16.137, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 5.2231526104417676e-06, | |
| "loss": 0.7113, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 0.6939824819564819, | |
| "eval_runtime": 64.7414, | |
| "eval_samples_per_second": 1544.606, | |
| "eval_steps_per_second": 16.095, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 5.022449799196788e-06, | |
| "loss": 0.7087, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_loss": 0.6922757029533386, | |
| "eval_runtime": 64.5094, | |
| "eval_samples_per_second": 1550.162, | |
| "eval_steps_per_second": 16.153, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.821746987951808e-06, | |
| "loss": 0.7071, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.6915081739425659, | |
| "eval_runtime": 65.0605, | |
| "eval_samples_per_second": 1537.031, | |
| "eval_steps_per_second": 16.016, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 4.6210441767068274e-06, | |
| "loss": 0.7047, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 0.6870027184486389, | |
| "eval_runtime": 64.7199, | |
| "eval_samples_per_second": 1545.119, | |
| "eval_steps_per_second": 16.1, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 4.420341365461848e-06, | |
| "loss": 0.7027, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_loss": 0.6844470500946045, | |
| "eval_runtime": 64.906, | |
| "eval_samples_per_second": 1540.69, | |
| "eval_steps_per_second": 16.054, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.219638554216868e-06, | |
| "loss": 0.6999, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 0.6833365559577942, | |
| "eval_runtime": 64.5739, | |
| "eval_samples_per_second": 1548.614, | |
| "eval_steps_per_second": 16.137, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 4.018935742971888e-06, | |
| "loss": 0.6992, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_loss": 0.6834575533866882, | |
| "eval_runtime": 64.6706, | |
| "eval_samples_per_second": 1546.298, | |
| "eval_steps_per_second": 16.112, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 3.818253012048193e-06, | |
| "loss": 0.6975, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_loss": 0.6793897151947021, | |
| "eval_runtime": 64.8495, | |
| "eval_samples_per_second": 1542.031, | |
| "eval_steps_per_second": 16.068, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 3.617530120481928e-06, | |
| "loss": 0.6947, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_loss": 0.6794592142105103, | |
| "eval_runtime": 64.6554, | |
| "eval_samples_per_second": 1546.661, | |
| "eval_steps_per_second": 16.116, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.416827309236948e-06, | |
| "loss": 0.6947, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_loss": 0.6781629920005798, | |
| "eval_runtime": 65.93, | |
| "eval_samples_per_second": 1516.76, | |
| "eval_steps_per_second": 15.805, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.216124497991968e-06, | |
| "loss": 0.6932, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.6771513819694519, | |
| "eval_runtime": 64.7971, | |
| "eval_samples_per_second": 1543.279, | |
| "eval_steps_per_second": 16.081, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 3.015401606425703e-06, | |
| "loss": 0.6914, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "eval_loss": 0.6757428050041199, | |
| "eval_runtime": 64.8856, | |
| "eval_samples_per_second": 1541.173, | |
| "eval_steps_per_second": 16.059, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.8146987951807233e-06, | |
| "loss": 0.6907, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_loss": 0.6730121374130249, | |
| "eval_runtime": 65.1098, | |
| "eval_samples_per_second": 1535.867, | |
| "eval_steps_per_second": 16.004, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 2.6139959839357434e-06, | |
| "loss": 0.689, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_loss": 0.6724720597267151, | |
| "eval_runtime": 65.1981, | |
| "eval_samples_per_second": 1533.787, | |
| "eval_steps_per_second": 15.982, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 2.413293172690763e-06, | |
| "loss": 0.6885, | |
| "step": 380000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "eval_loss": 0.6744215488433838, | |
| "eval_runtime": 64.9859, | |
| "eval_samples_per_second": 1538.796, | |
| "eval_steps_per_second": 16.034, | |
| "step": 380000 | |
| } | |
| ], | |
| "logging_steps": 10000, | |
| "max_steps": 500000, | |
| "num_train_epochs": 5, | |
| "save_steps": 10000, | |
| "total_flos": 6.301004166565632e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |