{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 6.0, "eval_steps": 500, "global_step": 48, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.125, "grad_norm": 29.546459197998047, "learning_rate": 0.0, "loss": 2.9823, "step": 1 }, { "epoch": 0.25, "grad_norm": 25.17301368713379, "learning_rate": 4e-05, "loss": 3.2575, "step": 2 }, { "epoch": 0.375, "grad_norm": 17.487424850463867, "learning_rate": 8e-05, "loss": 2.8881, "step": 3 }, { "epoch": 0.5, "grad_norm": 21.911840438842773, "learning_rate": 0.00012, "loss": 2.7858, "step": 4 }, { "epoch": 0.625, "grad_norm": 17.531702041625977, "learning_rate": 0.00016, "loss": 2.9163, "step": 5 }, { "epoch": 0.75, "grad_norm": 59.228965759277344, "learning_rate": 0.0002, "loss": 2.6071, "step": 6 }, { "epoch": 0.875, "grad_norm": 26.574459075927734, "learning_rate": 0.00019534883720930232, "loss": 2.6084, "step": 7 }, { "epoch": 1.0, "grad_norm": 21.419315338134766, "learning_rate": 0.00019069767441860466, "loss": 2.4804, "step": 8 }, { "epoch": 1.125, "grad_norm": 10.971414566040039, "learning_rate": 0.000186046511627907, "loss": 2.3148, "step": 9 }, { "epoch": 1.25, "grad_norm": 9.723897933959961, "learning_rate": 0.0001813953488372093, "loss": 2.0739, "step": 10 }, { "epoch": 1.375, "grad_norm": 10.31385612487793, "learning_rate": 0.00017674418604651164, "loss": 2.1645, "step": 11 }, { "epoch": 1.5, "grad_norm": 8.488858222961426, "learning_rate": 0.00017209302325581395, "loss": 1.9289, "step": 12 }, { "epoch": 1.625, "grad_norm": 8.55744457244873, "learning_rate": 0.00016744186046511629, "loss": 2.1265, "step": 13 }, { "epoch": 1.75, "grad_norm": 6.713319301605225, "learning_rate": 0.00016279069767441862, "loss": 2.0431, "step": 14 }, { "epoch": 1.875, "grad_norm": 5.568178176879883, "learning_rate": 0.00015813953488372093, "loss": 1.9469, "step": 15 }, { "epoch": 2.0, "grad_norm": 6.652850151062012, "learning_rate": 0.00015348837209302327, "loss": 2.0207, "step": 16 }, { "epoch": 2.125, "grad_norm": 4.384023189544678, "learning_rate": 0.00014883720930232558, "loss": 1.5723, "step": 17 }, { "epoch": 2.25, "grad_norm": 4.638486862182617, "learning_rate": 0.00014418604651162791, "loss": 1.8366, "step": 18 }, { "epoch": 2.375, "grad_norm": 4.115764617919922, "learning_rate": 0.00013953488372093025, "loss": 1.7704, "step": 19 }, { "epoch": 2.5, "grad_norm": 4.664848804473877, "learning_rate": 0.00013488372093023256, "loss": 1.8609, "step": 20 }, { "epoch": 2.625, "grad_norm": 4.360297679901123, "learning_rate": 0.0001302325581395349, "loss": 1.5956, "step": 21 }, { "epoch": 2.75, "grad_norm": 4.980749607086182, "learning_rate": 0.0001255813953488372, "loss": 1.707, "step": 22 }, { "epoch": 2.875, "grad_norm": 3.9513840675354004, "learning_rate": 0.00012093023255813953, "loss": 1.6126, "step": 23 }, { "epoch": 3.0, "grad_norm": 4.385582447052002, "learning_rate": 0.00011627906976744187, "loss": 1.9308, "step": 24 }, { "epoch": 3.125, "grad_norm": 3.52531099319458, "learning_rate": 0.00011162790697674419, "loss": 1.3958, "step": 25 }, { "epoch": 3.25, "grad_norm": 3.785879135131836, "learning_rate": 0.00010697674418604651, "loss": 1.6508, "step": 26 }, { "epoch": 3.375, "grad_norm": 3.628448486328125, "learning_rate": 0.00010232558139534885, "loss": 1.369, "step": 27 }, { "epoch": 3.5, "grad_norm": 3.9527487754821777, "learning_rate": 9.767441860465116e-05, "loss": 1.6467, "step": 28 }, { "epoch": 3.625, "grad_norm": 3.7579798698425293, "learning_rate": 9.30232558139535e-05, "loss": 1.523, "step": 29 }, { "epoch": 3.75, "grad_norm": 4.251528739929199, "learning_rate": 8.837209302325582e-05, "loss": 1.5922, "step": 30 }, { "epoch": 3.875, "grad_norm": 4.362473011016846, "learning_rate": 8.372093023255814e-05, "loss": 1.7363, "step": 31 }, { "epoch": 4.0, "grad_norm": 3.8354411125183105, "learning_rate": 7.906976744186047e-05, "loss": 1.4386, "step": 32 }, { "epoch": 4.125, "grad_norm": 3.2613353729248047, "learning_rate": 7.441860465116279e-05, "loss": 1.3475, "step": 33 }, { "epoch": 4.25, "grad_norm": 3.43796968460083, "learning_rate": 6.976744186046513e-05, "loss": 1.6003, "step": 34 }, { "epoch": 4.375, "grad_norm": 3.6074087619781494, "learning_rate": 6.511627906976745e-05, "loss": 1.3937, "step": 35 }, { "epoch": 4.5, "grad_norm": 3.9121947288513184, "learning_rate": 6.0465116279069765e-05, "loss": 1.4014, "step": 36 }, { "epoch": 4.625, "grad_norm": 3.7435214519500732, "learning_rate": 5.5813953488372095e-05, "loss": 1.407, "step": 37 }, { "epoch": 4.75, "grad_norm": 3.863990068435669, "learning_rate": 5.1162790697674425e-05, "loss": 1.443, "step": 38 }, { "epoch": 4.875, "grad_norm": 3.776215076446533, "learning_rate": 4.651162790697675e-05, "loss": 1.3739, "step": 39 }, { "epoch": 5.0, "grad_norm": 3.557727098464966, "learning_rate": 4.186046511627907e-05, "loss": 1.265, "step": 40 }, { "epoch": 5.125, "grad_norm": 3.0958468914031982, "learning_rate": 3.7209302325581394e-05, "loss": 1.3596, "step": 41 }, { "epoch": 5.25, "grad_norm": 3.090500593185425, "learning_rate": 3.2558139534883724e-05, "loss": 1.1442, "step": 42 }, { "epoch": 5.375, "grad_norm": 3.522876024246216, "learning_rate": 2.7906976744186048e-05, "loss": 1.4195, "step": 43 }, { "epoch": 5.5, "grad_norm": 3.2996609210968018, "learning_rate": 2.3255813953488374e-05, "loss": 1.4001, "step": 44 }, { "epoch": 5.625, "grad_norm": 3.25895094871521, "learning_rate": 1.8604651162790697e-05, "loss": 1.4081, "step": 45 }, { "epoch": 5.75, "grad_norm": 3.050686836242676, "learning_rate": 1.3953488372093024e-05, "loss": 1.1863, "step": 46 }, { "epoch": 5.875, "grad_norm": 3.598515033721924, "learning_rate": 9.302325581395349e-06, "loss": 1.4456, "step": 47 }, { "epoch": 6.0, "grad_norm": 3.6872923374176025, "learning_rate": 4.651162790697674e-06, "loss": 1.1993, "step": 48 } ], "logging_steps": 1, "max_steps": 48, "num_input_tokens_seen": 0, "num_train_epochs": 6, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 4.606095647108448e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }