{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "eval_steps": 21, "global_step": 84, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.023809523809523808, "grad_norm": 0.09562449157238007, "learning_rate": 0.0, "loss": 0.9296531081199646, "step": 1 }, { "epoch": 0.047619047619047616, "grad_norm": 0.08915263414382935, "learning_rate": 4e-05, "loss": 0.8963991403579712, "step": 2 }, { "epoch": 0.07142857142857142, "grad_norm": 0.09575905650854111, "learning_rate": 8e-05, "loss": 0.9559247493743896, "step": 3 }, { "epoch": 0.09523809523809523, "grad_norm": 0.09761863201856613, "learning_rate": 0.00012, "loss": 0.9154826402664185, "step": 4 }, { "epoch": 0.11904761904761904, "grad_norm": 0.11157004535198212, "learning_rate": 0.00016, "loss": 0.904577374458313, "step": 5 }, { "epoch": 0.14285714285714285, "grad_norm": 0.13675838708877563, "learning_rate": 0.0002, "loss": 0.8809080123901367, "step": 6 }, { "epoch": 0.16666666666666666, "grad_norm": 0.1358879655599594, "learning_rate": 0.00019834710743801655, "loss": 0.8010007739067078, "step": 7 }, { "epoch": 0.19047619047619047, "grad_norm": 0.1459621638059616, "learning_rate": 0.0001966942148760331, "loss": 0.7688196897506714, "step": 8 }, { "epoch": 0.21428571428571427, "grad_norm": 0.15010619163513184, "learning_rate": 0.0001950413223140496, "loss": 0.682942271232605, "step": 9 }, { "epoch": 0.23809523809523808, "grad_norm": 0.15406596660614014, "learning_rate": 0.0001933884297520661, "loss": 0.6647316217422485, "step": 10 }, { "epoch": 0.2619047619047619, "grad_norm": 0.165545254945755, "learning_rate": 0.00019173553719008265, "loss": 0.5688892006874084, "step": 11 }, { "epoch": 0.2857142857142857, "grad_norm": 0.17254289984703064, "learning_rate": 0.0001900826446280992, "loss": 0.5390745401382446, "step": 12 }, { "epoch": 0.30952380952380953, "grad_norm": 0.1956062614917755, "learning_rate": 0.0001884297520661157, "loss": 0.46194905042648315, "step": 13 }, { "epoch": 0.3333333333333333, "grad_norm": 0.19603870809078217, "learning_rate": 0.00018677685950413224, "loss": 0.41473817825317383, "step": 14 }, { "epoch": 0.35714285714285715, "grad_norm": 0.20052343606948853, "learning_rate": 0.00018512396694214878, "loss": 0.3571416139602661, "step": 15 }, { "epoch": 0.38095238095238093, "grad_norm": 0.24088889360427856, "learning_rate": 0.00018347107438016532, "loss": 0.37838655710220337, "step": 16 }, { "epoch": 0.40476190476190477, "grad_norm": 0.23278626799583435, "learning_rate": 0.00018181818181818183, "loss": 0.31288638710975647, "step": 17 }, { "epoch": 0.42857142857142855, "grad_norm": 0.18987713754177094, "learning_rate": 0.00018016528925619835, "loss": 0.25435182452201843, "step": 18 }, { "epoch": 0.4523809523809524, "grad_norm": 0.17959612607955933, "learning_rate": 0.00017851239669421489, "loss": 0.25271108746528625, "step": 19 }, { "epoch": 0.47619047619047616, "grad_norm": 0.16697828471660614, "learning_rate": 0.00017685950413223143, "loss": 0.23363591730594635, "step": 20 }, { "epoch": 0.5, "grad_norm": 0.14224942028522491, "learning_rate": 0.00017520661157024794, "loss": 0.20100688934326172, "step": 21 }, { "epoch": 0.5, "eval_loss": 0.23801279067993164, "eval_runtime": 11.6604, "eval_samples_per_second": 4.116, "eval_steps_per_second": 2.058, "step": 21 }, { "epoch": 0.5238095238095238, "grad_norm": 0.1497686207294464, "learning_rate": 0.00017355371900826448, "loss": 0.1805773675441742, "step": 22 }, { "epoch": 0.5476190476190477, "grad_norm": 0.13744786381721497, "learning_rate": 0.00017190082644628102, "loss": 0.15929169952869415, "step": 23 }, { "epoch": 0.5714285714285714, "grad_norm": 0.16764526069164276, "learning_rate": 0.00017024793388429753, "loss": 0.2225639373064041, "step": 24 }, { "epoch": 0.5952380952380952, "grad_norm": 0.17506416141986847, "learning_rate": 0.00016859504132231404, "loss": 0.18895968794822693, "step": 25 }, { "epoch": 0.6190476190476191, "grad_norm": 0.17485980689525604, "learning_rate": 0.00016694214876033058, "loss": 0.194553405046463, "step": 26 }, { "epoch": 0.6428571428571429, "grad_norm": 0.15143030881881714, "learning_rate": 0.00016528925619834712, "loss": 0.2037380486726761, "step": 27 }, { "epoch": 0.6666666666666666, "grad_norm": 0.15532037615776062, "learning_rate": 0.00016363636363636366, "loss": 0.15463697910308838, "step": 28 }, { "epoch": 0.6904761904761905, "grad_norm": 0.17867055535316467, "learning_rate": 0.00016198347107438017, "loss": 0.2169087827205658, "step": 29 }, { "epoch": 0.7142857142857143, "grad_norm": 0.1316670924425125, "learning_rate": 0.0001603305785123967, "loss": 0.14843176305294037, "step": 30 }, { "epoch": 0.7380952380952381, "grad_norm": 0.13806650042533875, "learning_rate": 0.00015867768595041322, "loss": 0.14517749845981598, "step": 31 }, { "epoch": 0.7619047619047619, "grad_norm": 0.13012579083442688, "learning_rate": 0.00015702479338842976, "loss": 0.15764746069908142, "step": 32 }, { "epoch": 0.7857142857142857, "grad_norm": 0.1331578642129898, "learning_rate": 0.00015537190082644627, "loss": 0.14381033182144165, "step": 33 }, { "epoch": 0.8095238095238095, "grad_norm": 0.13435214757919312, "learning_rate": 0.00015371900826446281, "loss": 0.18524238467216492, "step": 34 }, { "epoch": 0.8333333333333334, "grad_norm": 0.12382369488477707, "learning_rate": 0.00015206611570247935, "loss": 0.14671051502227783, "step": 35 }, { "epoch": 0.8571428571428571, "grad_norm": 0.13217748701572418, "learning_rate": 0.0001504132231404959, "loss": 0.13472357392311096, "step": 36 }, { "epoch": 0.8809523809523809, "grad_norm": 0.1581868827342987, "learning_rate": 0.0001487603305785124, "loss": 0.13790392875671387, "step": 37 }, { "epoch": 0.9047619047619048, "grad_norm": 0.15799134969711304, "learning_rate": 0.00014710743801652894, "loss": 0.19304180145263672, "step": 38 }, { "epoch": 0.9285714285714286, "grad_norm": 0.139670729637146, "learning_rate": 0.00014545454545454546, "loss": 0.1486513316631317, "step": 39 }, { "epoch": 0.9523809523809523, "grad_norm": 0.12580491602420807, "learning_rate": 0.000143801652892562, "loss": 0.12202717363834381, "step": 40 }, { "epoch": 0.9761904761904762, "grad_norm": 0.14996661245822906, "learning_rate": 0.0001421487603305785, "loss": 0.16839513182640076, "step": 41 }, { "epoch": 1.0, "grad_norm": 0.10198456048965454, "learning_rate": 0.00014049586776859505, "loss": 0.1132611632347107, "step": 42 }, { "epoch": 1.0, "eval_loss": 0.16770125925540924, "eval_runtime": 11.5284, "eval_samples_per_second": 4.164, "eval_steps_per_second": 2.082, "step": 42 }, { "epoch": 1.0238095238095237, "grad_norm": 0.1118578240275383, "learning_rate": 0.0001388429752066116, "loss": 0.12670840322971344, "step": 43 }, { "epoch": 1.0476190476190477, "grad_norm": 0.11127142608165741, "learning_rate": 0.00013719008264462813, "loss": 0.13318225741386414, "step": 44 }, { "epoch": 1.0714285714285714, "grad_norm": 0.11384990811347961, "learning_rate": 0.00013553719008264464, "loss": 0.14756664633750916, "step": 45 }, { "epoch": 1.0952380952380953, "grad_norm": 0.11951878666877747, "learning_rate": 0.00013388429752066115, "loss": 0.16569679975509644, "step": 46 }, { "epoch": 1.119047619047619, "grad_norm": 0.09811879694461823, "learning_rate": 0.0001322314049586777, "loss": 0.1144881546497345, "step": 47 }, { "epoch": 1.1428571428571428, "grad_norm": 0.08161773532629013, "learning_rate": 0.00013057851239669423, "loss": 0.11602826416492462, "step": 48 }, { "epoch": 1.1666666666666667, "grad_norm": 0.11749138683080673, "learning_rate": 0.00012892561983471074, "loss": 0.15802887082099915, "step": 49 }, { "epoch": 1.1904761904761905, "grad_norm": 0.0926203802227974, "learning_rate": 0.00012727272727272728, "loss": 0.11292621493339539, "step": 50 }, { "epoch": 1.2142857142857142, "grad_norm": 0.1123381108045578, "learning_rate": 0.00012561983471074382, "loss": 0.14036691188812256, "step": 51 }, { "epoch": 1.2380952380952381, "grad_norm": 0.09942558407783508, "learning_rate": 0.00012396694214876033, "loss": 0.11614295095205307, "step": 52 }, { "epoch": 1.2619047619047619, "grad_norm": 0.10968922078609467, "learning_rate": 0.00012231404958677685, "loss": 0.17394062876701355, "step": 53 }, { "epoch": 1.2857142857142856, "grad_norm": 0.1058293879032135, "learning_rate": 0.0001206611570247934, "loss": 0.16215264797210693, "step": 54 }, { "epoch": 1.3095238095238095, "grad_norm": 0.09397002309560776, "learning_rate": 0.00011900826446280992, "loss": 0.10367283225059509, "step": 55 }, { "epoch": 1.3333333333333333, "grad_norm": 0.11522696167230606, "learning_rate": 0.00011735537190082646, "loss": 0.11446119844913483, "step": 56 }, { "epoch": 1.3571428571428572, "grad_norm": 0.12047811597585678, "learning_rate": 0.00011570247933884298, "loss": 0.10753891617059708, "step": 57 }, { "epoch": 1.380952380952381, "grad_norm": 0.14889945089817047, "learning_rate": 0.0001140495867768595, "loss": 0.15190932154655457, "step": 58 }, { "epoch": 1.4047619047619047, "grad_norm": 0.10881127417087555, "learning_rate": 0.00011239669421487604, "loss": 0.14703251421451569, "step": 59 }, { "epoch": 1.4285714285714286, "grad_norm": 0.085318922996521, "learning_rate": 0.00011074380165289258, "loss": 0.08669870346784592, "step": 60 }, { "epoch": 1.4523809523809523, "grad_norm": 0.10852408409118652, "learning_rate": 0.00010909090909090909, "loss": 0.1309637427330017, "step": 61 }, { "epoch": 1.4761904761904763, "grad_norm": 0.17673085629940033, "learning_rate": 0.00010743801652892562, "loss": 0.13333159685134888, "step": 62 }, { "epoch": 1.5, "grad_norm": 0.07902995496988297, "learning_rate": 0.00010578512396694216, "loss": 0.0821666419506073, "step": 63 }, { "epoch": 1.5, "eval_loss": 0.16188834607601166, "eval_runtime": 11.4102, "eval_samples_per_second": 4.207, "eval_steps_per_second": 2.103, "step": 63 }, { "epoch": 1.5238095238095237, "grad_norm": 0.0994923785328865, "learning_rate": 0.0001041322314049587, "loss": 0.12175415456295013, "step": 64 }, { "epoch": 1.5476190476190477, "grad_norm": 0.10065561532974243, "learning_rate": 0.00010247933884297521, "loss": 0.11586101353168488, "step": 65 }, { "epoch": 1.5714285714285714, "grad_norm": 0.10240356624126434, "learning_rate": 0.00010082644628099174, "loss": 0.09786738455295563, "step": 66 }, { "epoch": 1.5952380952380953, "grad_norm": 0.11477634310722351, "learning_rate": 9.917355371900827e-05, "loss": 0.12858846783638, "step": 67 }, { "epoch": 1.619047619047619, "grad_norm": 0.13524135947227478, "learning_rate": 9.75206611570248e-05, "loss": 0.1482604295015335, "step": 68 }, { "epoch": 1.6428571428571428, "grad_norm": 0.11171480268239975, "learning_rate": 9.586776859504133e-05, "loss": 0.1510271579027176, "step": 69 }, { "epoch": 1.6666666666666665, "grad_norm": 0.11093535274267197, "learning_rate": 9.421487603305785e-05, "loss": 0.17145098745822906, "step": 70 }, { "epoch": 1.6904761904761905, "grad_norm": 0.09573192894458771, "learning_rate": 9.256198347107439e-05, "loss": 0.12275607883930206, "step": 71 }, { "epoch": 1.7142857142857144, "grad_norm": 0.08978857845067978, "learning_rate": 9.090909090909092e-05, "loss": 0.0835709348320961, "step": 72 }, { "epoch": 1.7380952380952381, "grad_norm": 0.10285310447216034, "learning_rate": 8.925619834710744e-05, "loss": 0.140127494931221, "step": 73 }, { "epoch": 1.7619047619047619, "grad_norm": 0.11121945083141327, "learning_rate": 8.760330578512397e-05, "loss": 0.15469461679458618, "step": 74 }, { "epoch": 1.7857142857142856, "grad_norm": 0.0986233502626419, "learning_rate": 8.595041322314051e-05, "loss": 0.13605977594852448, "step": 75 }, { "epoch": 1.8095238095238095, "grad_norm": 0.10802352428436279, "learning_rate": 8.429752066115702e-05, "loss": 0.1513635218143463, "step": 76 }, { "epoch": 1.8333333333333335, "grad_norm": 0.10008759051561356, "learning_rate": 8.264462809917356e-05, "loss": 0.12034285813570023, "step": 77 }, { "epoch": 1.8571428571428572, "grad_norm": 0.100432388484478, "learning_rate": 8.099173553719009e-05, "loss": 0.13318830728530884, "step": 78 }, { "epoch": 1.880952380952381, "grad_norm": 0.13922077417373657, "learning_rate": 7.933884297520661e-05, "loss": 0.13727393746376038, "step": 79 }, { "epoch": 1.9047619047619047, "grad_norm": 0.09379437565803528, "learning_rate": 7.768595041322314e-05, "loss": 0.11401292681694031, "step": 80 }, { "epoch": 1.9285714285714286, "grad_norm": 0.09079214930534363, "learning_rate": 7.603305785123968e-05, "loss": 0.11084987223148346, "step": 81 }, { "epoch": 1.9523809523809523, "grad_norm": 0.09683443605899811, "learning_rate": 7.43801652892562e-05, "loss": 0.11253248155117035, "step": 82 }, { "epoch": 1.9761904761904763, "grad_norm": 0.10992921888828278, "learning_rate": 7.272727272727273e-05, "loss": 0.15756897628307343, "step": 83 }, { "epoch": 2.0, "grad_norm": 0.12596677243709564, "learning_rate": 7.107438016528925e-05, "loss": 0.15924668312072754, "step": 84 }, { "epoch": 2.0, "eval_loss": 0.15902404487133026, "eval_runtime": 11.6031, "eval_samples_per_second": 4.137, "eval_steps_per_second": 2.068, "step": 84 } ], "logging_steps": 1, "max_steps": 126, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 21, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2.9359625119727616e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }