| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.578947368421053, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0002962686567164179, | |
| "loss": 1.5814, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.0006077612633816898, | |
| "eval_runtime": 128.8821, | |
| "eval_samples_per_second": 75.48, | |
| "eval_steps_per_second": 2.359, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00029004975124378106, | |
| "loss": 0.0051, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 0.0004873188154306263, | |
| "eval_runtime": 129.1791, | |
| "eval_samples_per_second": 75.306, | |
| "eval_steps_per_second": 2.353, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00028383084577114425, | |
| "loss": 0.0033, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 0.0002883475972339511, | |
| "eval_runtime": 128.9078, | |
| "eval_samples_per_second": 75.465, | |
| "eval_steps_per_second": 2.358, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00027761194029850744, | |
| "loss": 0.0026, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 0.00016567455895710737, | |
| "eval_runtime": 128.8984, | |
| "eval_samples_per_second": 75.47, | |
| "eval_steps_per_second": 2.358, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00027139303482587063, | |
| "loss": 0.002, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.0001291305961785838, | |
| "eval_runtime": 128.8976, | |
| "eval_samples_per_second": 75.471, | |
| "eval_steps_per_second": 2.358, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0002651741293532338, | |
| "loss": 0.0016, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 0.00010879799810936674, | |
| "eval_runtime": 128.8894, | |
| "eval_samples_per_second": 75.476, | |
| "eval_steps_per_second": 2.359, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.000258955223880597, | |
| "loss": 0.0014, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": 0.00010257180110784248, | |
| "eval_runtime": 129.1985, | |
| "eval_samples_per_second": 75.295, | |
| "eval_steps_per_second": 2.353, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.0002527363184079602, | |
| "loss": 0.0012, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_loss": 6.881527951918542e-05, | |
| "eval_runtime": 129.1934, | |
| "eval_samples_per_second": 75.298, | |
| "eval_steps_per_second": 2.353, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00024651741293532333, | |
| "loss": 0.0016, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 7.058009214233607e-05, | |
| "eval_runtime": 128.8993, | |
| "eval_samples_per_second": 75.47, | |
| "eval_steps_per_second": 2.358, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.00024029850746268655, | |
| "loss": 0.001, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 6.400555139407516e-05, | |
| "eval_runtime": 128.8789, | |
| "eval_samples_per_second": 75.482, | |
| "eval_steps_per_second": 2.359, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002340796019900497, | |
| "loss": 0.001, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 6.463182216975838e-05, | |
| "eval_runtime": 128.8854, | |
| "eval_samples_per_second": 75.478, | |
| "eval_steps_per_second": 2.359, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00022786069651741292, | |
| "loss": 0.0012, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.0001349268713966012, | |
| "eval_runtime": 128.8866, | |
| "eval_samples_per_second": 75.477, | |
| "eval_steps_per_second": 2.359, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002216417910447761, | |
| "loss": 0.0009, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 6.4761312387418e-05, | |
| "eval_runtime": 129.1959, | |
| "eval_samples_per_second": 75.296, | |
| "eval_steps_per_second": 2.353, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00021542288557213927, | |
| "loss": 0.0009, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 5.111826249049045e-05, | |
| "eval_runtime": 128.8809, | |
| "eval_samples_per_second": 75.481, | |
| "eval_steps_per_second": 2.359, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.00020920398009950246, | |
| "loss": 0.0007, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_loss": 6.521103205159307e-05, | |
| "eval_runtime": 128.9, | |
| "eval_samples_per_second": 75.469, | |
| "eval_steps_per_second": 2.358, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 0.00020298507462686565, | |
| "loss": 0.0007, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_loss": 4.892368451692164e-05, | |
| "eval_runtime": 128.8971, | |
| "eval_samples_per_second": 75.471, | |
| "eval_steps_per_second": 2.358, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.00019676616915422884, | |
| "loss": 0.0006, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_loss": 4.084123429493047e-05, | |
| "eval_runtime": 69.0191, | |
| "eval_samples_per_second": 140.947, | |
| "eval_steps_per_second": 4.405, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.00019054726368159203, | |
| "loss": 0.0006, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 6.229727296158671e-05, | |
| "eval_runtime": 83.9466, | |
| "eval_samples_per_second": 115.883, | |
| "eval_steps_per_second": 3.621, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.00018432835820895522, | |
| "loss": 0.0008, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_loss": 4.437619645614177e-05, | |
| "eval_runtime": 129.1502, | |
| "eval_samples_per_second": 75.323, | |
| "eval_steps_per_second": 2.354, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.00017810945273631838, | |
| "loss": 0.0008, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "eval_loss": 4.2610481614246964e-05, | |
| "eval_runtime": 128.7805, | |
| "eval_samples_per_second": 75.539, | |
| "eval_steps_per_second": 2.361, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.00017189054726368157, | |
| "loss": 0.0005, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "eval_loss": 7.266196189448237e-05, | |
| "eval_runtime": 128.7609, | |
| "eval_samples_per_second": 75.551, | |
| "eval_steps_per_second": 2.361, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.00016567164179104478, | |
| "loss": 0.0006, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "eval_loss": 3.264641054556705e-05, | |
| "eval_runtime": 128.7913, | |
| "eval_samples_per_second": 75.533, | |
| "eval_steps_per_second": 2.36, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.00015945273631840794, | |
| "loss": 0.0005, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "eval_loss": 2.6435563995619304e-05, | |
| "eval_runtime": 128.7891, | |
| "eval_samples_per_second": 75.534, | |
| "eval_steps_per_second": 2.36, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 0.00015323383084577113, | |
| "loss": 0.0006, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "eval_loss": 4.4951906602364033e-05, | |
| "eval_runtime": 129.1622, | |
| "eval_samples_per_second": 75.316, | |
| "eval_steps_per_second": 2.354, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.00014701492537313432, | |
| "loss": 0.0005, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_loss": 3.5103537811664864e-05, | |
| "eval_runtime": 128.7717, | |
| "eval_samples_per_second": 75.545, | |
| "eval_steps_per_second": 2.361, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 0.00014079601990049748, | |
| "loss": 0.0004, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "eval_loss": 3.251100133638829e-05, | |
| "eval_runtime": 128.8099, | |
| "eval_samples_per_second": 75.522, | |
| "eval_steps_per_second": 2.36, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 0.0001345771144278607, | |
| "loss": 0.0004, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "eval_loss": 3.5094544728053734e-05, | |
| "eval_runtime": 128.8083, | |
| "eval_samples_per_second": 75.523, | |
| "eval_steps_per_second": 2.36, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.00012835820895522386, | |
| "loss": 0.0005, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_loss": 3.747216032934375e-05, | |
| "eval_runtime": 128.7944, | |
| "eval_samples_per_second": 75.531, | |
| "eval_steps_per_second": 2.36, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.00012213930348258705, | |
| "loss": 0.0004, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "eval_loss": 2.8431584723875858e-05, | |
| "eval_runtime": 129.1362, | |
| "eval_samples_per_second": 75.331, | |
| "eval_steps_per_second": 2.354, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.00011592039800995025, | |
| "loss": 0.0003, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "eval_loss": 2.733626024564728e-05, | |
| "eval_runtime": 128.9297, | |
| "eval_samples_per_second": 75.452, | |
| "eval_steps_per_second": 2.358, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.00010970149253731342, | |
| "loss": 0.0004, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "eval_loss": 2.5848928999039344e-05, | |
| "eval_runtime": 128.8142, | |
| "eval_samples_per_second": 75.52, | |
| "eval_steps_per_second": 2.36, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 0.0001034825870646766, | |
| "loss": 0.0003, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "eval_loss": 3.0041233912925236e-05, | |
| "eval_runtime": 128.8146, | |
| "eval_samples_per_second": 75.519, | |
| "eval_steps_per_second": 2.36, | |
| "step": 8000 | |
| } | |
| ], | |
| "max_steps": 12160, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.757334967871078e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |