{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.9411764705882355, "eval_steps": 500, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.03, "learning_rate": 0.00019805941782534764, "loss": 2.0676, "step": 1 }, { "epoch": 0.06, "learning_rate": 0.00019605881764529358, "loss": 2.165, "step": 2 }, { "epoch": 0.09, "learning_rate": 0.00019405821746523957, "loss": 1.6457, "step": 3 }, { "epoch": 0.12, "learning_rate": 0.00019205761728518557, "loss": 1.2642, "step": 4 }, { "epoch": 0.15, "learning_rate": 0.00019005701710513156, "loss": 0.9945, "step": 5 }, { "epoch": 0.18, "learning_rate": 0.00018805641692507753, "loss": 1.1607, "step": 6 }, { "epoch": 0.21, "learning_rate": 0.00018605581674502352, "loss": 0.8384, "step": 7 }, { "epoch": 0.24, "learning_rate": 0.00018405521656496952, "loss": 1.4198, "step": 8 }, { "epoch": 0.26, "learning_rate": 0.00018205461638491548, "loss": 1.065, "step": 9 }, { "epoch": 0.29, "learning_rate": 0.00018005401620486148, "loss": 1.5488, "step": 10 }, { "epoch": 0.32, "learning_rate": 0.00017805341602480744, "loss": 0.926, "step": 11 }, { "epoch": 0.35, "learning_rate": 0.00017605281584475344, "loss": 1.1462, "step": 12 }, { "epoch": 0.38, "learning_rate": 0.00017405221566469943, "loss": 1.288, "step": 13 }, { "epoch": 0.41, "learning_rate": 0.00017205161548464542, "loss": 0.9373, "step": 14 }, { "epoch": 0.44, "learning_rate": 0.00017005101530459136, "loss": 0.7286, "step": 15 }, { "epoch": 0.47, "learning_rate": 0.00016805041512453736, "loss": 1.1179, "step": 16 }, { "epoch": 0.5, "learning_rate": 0.00016604981494448335, "loss": 0.6268, "step": 17 }, { "epoch": 0.53, "learning_rate": 0.00016404921476442935, "loss": 1.2474, "step": 18 }, { "epoch": 0.56, "learning_rate": 0.0001620486145843753, "loss": 0.8392, "step": 19 }, { "epoch": 0.59, "learning_rate": 0.0001600480144043213, "loss": 1.1382, "step": 20 }, { "epoch": 0.62, "learning_rate": 0.0001580474142242673, "loss": 1.171, "step": 21 }, { "epoch": 0.65, "learning_rate": 0.00015604681404421327, "loss": 0.8168, "step": 22 }, { "epoch": 0.68, "learning_rate": 0.00015404621386415926, "loss": 1.336, "step": 23 }, { "epoch": 0.71, "learning_rate": 0.00015204561368410523, "loss": 1.148, "step": 24 }, { "epoch": 0.74, "learning_rate": 0.00015004501350405122, "loss": 0.9818, "step": 25 }, { "epoch": 0.76, "learning_rate": 0.00014804441332399721, "loss": 1.1472, "step": 26 }, { "epoch": 0.79, "learning_rate": 0.0001460438131439432, "loss": 1.0053, "step": 27 }, { "epoch": 0.82, "learning_rate": 0.00014404321296388918, "loss": 1.053, "step": 28 }, { "epoch": 0.85, "learning_rate": 0.00014204261278383514, "loss": 1.2345, "step": 29 }, { "epoch": 0.88, "learning_rate": 0.00014004201260378114, "loss": 0.9872, "step": 30 }, { "epoch": 0.91, "learning_rate": 0.00013804141242372713, "loss": 1.1111, "step": 31 }, { "epoch": 0.94, "learning_rate": 0.00013604081224367312, "loss": 0.9571, "step": 32 }, { "epoch": 0.97, "learning_rate": 0.0001340402120636191, "loss": 1.0676, "step": 33 }, { "epoch": 1.0, "learning_rate": 0.00013203961188356508, "loss": 0.9105, "step": 34 }, { "epoch": 1.03, "learning_rate": 0.00013003901170351108, "loss": 0.638, "step": 35 }, { "epoch": 1.06, "learning_rate": 0.00012803841152345704, "loss": 1.0111, "step": 36 }, { "epoch": 1.09, "learning_rate": 0.000126037811343403, "loss": 0.6401, "step": 37 }, { "epoch": 1.12, "learning_rate": 0.000124037211163349, "loss": 0.7887, "step": 38 }, { "epoch": 1.15, "learning_rate": 0.000122036610983295, "loss": 0.7075, "step": 39 }, { "epoch": 1.18, "learning_rate": 0.00012003601080324098, "loss": 0.7454, "step": 40 }, { "epoch": 1.21, "learning_rate": 0.00011803541062318697, "loss": 0.9572, "step": 41 }, { "epoch": 1.24, "learning_rate": 0.00011603481044313295, "loss": 0.6438, "step": 42 }, { "epoch": 1.26, "learning_rate": 0.00011403421026307892, "loss": 0.6731, "step": 43 }, { "epoch": 1.29, "learning_rate": 0.00011203361008302491, "loss": 0.6977, "step": 44 }, { "epoch": 1.32, "learning_rate": 0.0001100330099029709, "loss": 0.6028, "step": 45 }, { "epoch": 1.35, "learning_rate": 0.00010803240972291689, "loss": 0.3236, "step": 46 }, { "epoch": 1.38, "learning_rate": 0.00010603180954286287, "loss": 0.7818, "step": 47 }, { "epoch": 1.41, "learning_rate": 0.00010403120936280886, "loss": 0.6282, "step": 48 }, { "epoch": 1.44, "learning_rate": 0.00010203060918275482, "loss": 0.7241, "step": 49 }, { "epoch": 1.47, "learning_rate": 0.00010003000900270081, "loss": 1.014, "step": 50 }, { "epoch": 1.5, "learning_rate": 9.802940882264679e-05, "loss": 0.5843, "step": 51 }, { "epoch": 1.53, "learning_rate": 9.602880864259278e-05, "loss": 0.5066, "step": 52 }, { "epoch": 1.56, "learning_rate": 9.402820846253876e-05, "loss": 0.922, "step": 53 }, { "epoch": 1.59, "learning_rate": 9.202760828248476e-05, "loss": 0.5464, "step": 54 }, { "epoch": 1.62, "learning_rate": 9.002700810243074e-05, "loss": 0.591, "step": 55 }, { "epoch": 1.65, "learning_rate": 8.802640792237672e-05, "loss": 0.8202, "step": 56 }, { "epoch": 1.68, "learning_rate": 8.602580774232271e-05, "loss": 0.8385, "step": 57 }, { "epoch": 1.71, "learning_rate": 8.402520756226868e-05, "loss": 0.8093, "step": 58 }, { "epoch": 1.74, "learning_rate": 8.202460738221467e-05, "loss": 0.7677, "step": 59 }, { "epoch": 1.76, "learning_rate": 8.002400720216065e-05, "loss": 0.7852, "step": 60 }, { "epoch": 1.79, "learning_rate": 7.802340702210663e-05, "loss": 0.9299, "step": 61 }, { "epoch": 1.82, "learning_rate": 7.602280684205261e-05, "loss": 0.7537, "step": 62 }, { "epoch": 1.85, "learning_rate": 7.402220666199861e-05, "loss": 0.8504, "step": 63 }, { "epoch": 1.88, "learning_rate": 7.202160648194459e-05, "loss": 0.6326, "step": 64 }, { "epoch": 1.91, "learning_rate": 7.002100630189057e-05, "loss": 0.4325, "step": 65 }, { "epoch": 1.94, "learning_rate": 6.802040612183656e-05, "loss": 0.7092, "step": 66 }, { "epoch": 1.97, "learning_rate": 6.601980594178254e-05, "loss": 0.9081, "step": 67 }, { "epoch": 2.0, "learning_rate": 6.401920576172852e-05, "loss": 0.5045, "step": 68 }, { "epoch": 2.03, "learning_rate": 6.20186055816745e-05, "loss": 0.418, "step": 69 }, { "epoch": 2.06, "learning_rate": 6.001800540162049e-05, "loss": 0.4485, "step": 70 }, { "epoch": 2.09, "learning_rate": 5.801740522156648e-05, "loss": 0.4887, "step": 71 }, { "epoch": 2.12, "learning_rate": 5.601680504151246e-05, "loss": 0.3122, "step": 72 }, { "epoch": 2.15, "learning_rate": 5.4016204861458444e-05, "loss": 0.4811, "step": 73 }, { "epoch": 2.18, "learning_rate": 5.201560468140443e-05, "loss": 0.2449, "step": 74 }, { "epoch": 2.21, "learning_rate": 5.0015004501350405e-05, "loss": 0.4157, "step": 75 }, { "epoch": 2.24, "learning_rate": 4.801440432129639e-05, "loss": 0.3642, "step": 76 }, { "epoch": 2.26, "learning_rate": 4.601380414124238e-05, "loss": 0.7396, "step": 77 }, { "epoch": 2.29, "learning_rate": 4.401320396118836e-05, "loss": 0.5625, "step": 78 }, { "epoch": 2.32, "learning_rate": 4.201260378113434e-05, "loss": 0.6648, "step": 79 }, { "epoch": 2.35, "learning_rate": 4.0012003601080326e-05, "loss": 0.3184, "step": 80 }, { "epoch": 2.38, "learning_rate": 3.801140342102631e-05, "loss": 0.3801, "step": 81 }, { "epoch": 2.41, "learning_rate": 3.6010803240972294e-05, "loss": 0.4922, "step": 82 }, { "epoch": 2.44, "learning_rate": 3.401020306091828e-05, "loss": 0.3716, "step": 83 }, { "epoch": 2.47, "learning_rate": 3.200960288086426e-05, "loss": 0.6029, "step": 84 }, { "epoch": 2.5, "learning_rate": 3.0009002700810245e-05, "loss": 0.5193, "step": 85 }, { "epoch": 2.53, "learning_rate": 2.800840252075623e-05, "loss": 0.5131, "step": 86 }, { "epoch": 2.56, "learning_rate": 2.6007802340702216e-05, "loss": 0.4021, "step": 87 }, { "epoch": 2.59, "learning_rate": 2.4007202160648196e-05, "loss": 0.1953, "step": 88 }, { "epoch": 2.62, "learning_rate": 2.200660198059418e-05, "loss": 0.4305, "step": 89 }, { "epoch": 2.65, "learning_rate": 2.0006001800540163e-05, "loss": 0.3563, "step": 90 }, { "epoch": 2.68, "learning_rate": 1.8005401620486147e-05, "loss": 0.4803, "step": 91 }, { "epoch": 2.71, "learning_rate": 1.600480144043213e-05, "loss": 0.3894, "step": 92 }, { "epoch": 2.74, "learning_rate": 1.4004201260378114e-05, "loss": 0.5841, "step": 93 }, { "epoch": 2.76, "learning_rate": 1.2003601080324098e-05, "loss": 0.2775, "step": 94 }, { "epoch": 2.79, "learning_rate": 1.0003000900270082e-05, "loss": 0.2276, "step": 95 }, { "epoch": 2.82, "learning_rate": 8.002400720216065e-06, "loss": 0.2473, "step": 96 }, { "epoch": 2.85, "learning_rate": 6.001800540162049e-06, "loss": 0.4294, "step": 97 }, { "epoch": 2.88, "learning_rate": 4.001200360108033e-06, "loss": 0.4284, "step": 98 }, { "epoch": 2.91, "learning_rate": 2.0006001800540163e-06, "loss": 0.2537, "step": 99 }, { "epoch": 2.94, "learning_rate": 0.0, "loss": 0.3062, "step": 100 } ], "logging_steps": 1, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 10, "total_flos": 6145255647191040.0, "train_batch_size": 1, "trial_name": null, "trial_params": null }