| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 17, | |
| "global_step": 84, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011904761904761904, | |
| "grad_norm": 0.26561033725738525, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 2.2541, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.023809523809523808, | |
| "grad_norm": 0.2501536011695862, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 2.2996, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03571428571428571, | |
| "grad_norm": 0.23626406490802765, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2516, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.047619047619047616, | |
| "grad_norm": 0.2351306676864624, | |
| "learning_rate": 0.00019992479525042303, | |
| "loss": 2.0972, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05952380952380952, | |
| "grad_norm": 0.29416894912719727, | |
| "learning_rate": 0.0001996992941167792, | |
| "loss": 1.9448, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "grad_norm": 0.4979971945285797, | |
| "learning_rate": 0.00019932383577419432, | |
| "loss": 1.8433, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "grad_norm": 0.3428308963775635, | |
| "learning_rate": 0.00019879898494768093, | |
| "loss": 1.7351, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 0.46657490730285645, | |
| "learning_rate": 0.00019812553106273847, | |
| "loss": 1.5867, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.10714285714285714, | |
| "grad_norm": 0.4063335955142975, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 1.3473, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.11904761904761904, | |
| "grad_norm": 0.5489273071289062, | |
| "learning_rate": 0.00019633708786158806, | |
| "loss": 1.3747, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13095238095238096, | |
| "grad_norm": 0.3660918176174164, | |
| "learning_rate": 0.00019522478853384155, | |
| "loss": 1.2575, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 0.28144609928131104, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 1.2127, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.15476190476190477, | |
| "grad_norm": 0.299657940864563, | |
| "learning_rate": 0.00019257239692688907, | |
| "loss": 1.2426, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 0.18204455077648163, | |
| "learning_rate": 0.0001910362940966147, | |
| "loss": 1.1797, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.17857142857142858, | |
| "grad_norm": 0.19023093581199646, | |
| "learning_rate": 0.00018936326403234125, | |
| "loss": 1.1511, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 0.12908512353897095, | |
| "learning_rate": 0.0001875558231302091, | |
| "loss": 1.0714, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.20238095238095238, | |
| "grad_norm": 0.10724656283855438, | |
| "learning_rate": 0.00018561668995302667, | |
| "loss": 1.1114, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.20238095238095238, | |
| "eval_loss": 1.1259114742279053, | |
| "eval_runtime": 248.1643, | |
| "eval_samples_per_second": 2.708, | |
| "eval_steps_per_second": 0.338, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 0.10084182024002075, | |
| "learning_rate": 0.00018354878114129367, | |
| "loss": 1.2143, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2261904761904762, | |
| "grad_norm": 0.14486701786518097, | |
| "learning_rate": 0.00018135520702629675, | |
| "loss": 1.127, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 0.08842574805021286, | |
| "learning_rate": 0.00017903926695187595, | |
| "loss": 1.1359, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.08982816338539124, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 1.136, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2619047619047619, | |
| "grad_norm": 0.08831178396940231, | |
| "learning_rate": 0.00017405440131090048, | |
| "loss": 1.104, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.27380952380952384, | |
| "grad_norm": 0.09078623354434967, | |
| "learning_rate": 0.00017139297345578994, | |
| "loss": 1.1852, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 0.09844227880239487, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 1.1274, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2976190476190476, | |
| "grad_norm": 0.09135384112596512, | |
| "learning_rate": 0.0001657521368569064, | |
| "loss": 1.1203, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.30952380952380953, | |
| "grad_norm": 0.09475894272327423, | |
| "learning_rate": 0.00016278121246720987, | |
| "loss": 1.0698, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.32142857142857145, | |
| "grad_norm": 0.09128192812204361, | |
| "learning_rate": 0.00015971585917027862, | |
| "loss": 1.1008, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.1022096574306488, | |
| "learning_rate": 0.00015656068754865387, | |
| "loss": 1.1174, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.34523809523809523, | |
| "grad_norm": 0.09190971404314041, | |
| "learning_rate": 0.00015332044328016914, | |
| "loss": 1.077, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 0.09310519695281982, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.1045, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.36904761904761907, | |
| "grad_norm": 0.09474918991327286, | |
| "learning_rate": 0.0001466043519702539, | |
| "loss": 1.1492, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 0.08711487054824829, | |
| "learning_rate": 0.00014313860656812536, | |
| "loss": 1.0322, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.39285714285714285, | |
| "grad_norm": 0.09338130801916122, | |
| "learning_rate": 0.0001396079766039157, | |
| "loss": 1.1269, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.40476190476190477, | |
| "grad_norm": 0.08810640871524811, | |
| "learning_rate": 0.00013601777248047105, | |
| "loss": 1.1308, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.40476190476190477, | |
| "eval_loss": 1.067408561706543, | |
| "eval_runtime": 248.1784, | |
| "eval_samples_per_second": 2.708, | |
| "eval_steps_per_second": 0.338, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 0.10047192871570587, | |
| "learning_rate": 0.00013237339420583212, | |
| "loss": 1.1319, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 0.10280368477106094, | |
| "learning_rate": 0.00012868032327110904, | |
| "loss": 1.1145, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.44047619047619047, | |
| "grad_norm": 0.0960673838853836, | |
| "learning_rate": 0.00012494411440579814, | |
| "loss": 1.0966, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.4523809523809524, | |
| "grad_norm": 0.10409148037433624, | |
| "learning_rate": 0.0001211703872229411, | |
| "loss": 1.0936, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.4642857142857143, | |
| "grad_norm": 0.09941337257623672, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.109, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 0.09119969606399536, | |
| "learning_rate": 0.00011353312997501313, | |
| "loss": 0.9867, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4880952380952381, | |
| "grad_norm": 0.09008549153804779, | |
| "learning_rate": 0.00010968108707031792, | |
| "loss": 1.0514, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.09608256071805954, | |
| "learning_rate": 0.00010581448289104758, | |
| "loss": 1.05, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5119047619047619, | |
| "grad_norm": 0.09066122770309448, | |
| "learning_rate": 0.00010193913317718244, | |
| "loss": 1.1273, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.5238095238095238, | |
| "grad_norm": 0.09085338562726974, | |
| "learning_rate": 9.806086682281758e-05, | |
| "loss": 1.0426, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 0.08922470360994339, | |
| "learning_rate": 9.418551710895243e-05, | |
| "loss": 1.0485, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5476190476190477, | |
| "grad_norm": 0.09132881462574005, | |
| "learning_rate": 9.03189129296821e-05, | |
| "loss": 1.0966, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5595238095238095, | |
| "grad_norm": 0.09102962166070938, | |
| "learning_rate": 8.646687002498692e-05, | |
| "loss": 1.0555, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.09371238946914673, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 1.0693, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5833333333333334, | |
| "grad_norm": 0.08540762215852737, | |
| "learning_rate": 7.882961277705895e-05, | |
| "loss": 0.9991, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "grad_norm": 0.08648107200860977, | |
| "learning_rate": 7.505588559420189e-05, | |
| "loss": 1.0399, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6071428571428571, | |
| "grad_norm": 0.08525101095438004, | |
| "learning_rate": 7.131967672889101e-05, | |
| "loss": 1.0251, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6071428571428571, | |
| "eval_loss": 1.0418401956558228, | |
| "eval_runtime": 248.0851, | |
| "eval_samples_per_second": 2.709, | |
| "eval_steps_per_second": 0.339, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6190476190476191, | |
| "grad_norm": 0.09358056634664536, | |
| "learning_rate": 6.762660579416791e-05, | |
| "loss": 1.151, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.6309523809523809, | |
| "grad_norm": 0.09469255059957504, | |
| "learning_rate": 6.398222751952899e-05, | |
| "loss": 1.0479, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.6428571428571429, | |
| "grad_norm": 0.09118176251649857, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 1.0467, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6547619047619048, | |
| "grad_norm": 0.0902942344546318, | |
| "learning_rate": 5.6861393431874675e-05, | |
| "loss": 0.9988, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.09111825376749039, | |
| "learning_rate": 5.339564802974615e-05, | |
| "loss": 1.0067, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.6785714285714286, | |
| "grad_norm": 0.1021014079451561, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 1.0492, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6904761904761905, | |
| "grad_norm": 0.09211082756519318, | |
| "learning_rate": 4.66795567198309e-05, | |
| "loss": 1.0085, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.7023809523809523, | |
| "grad_norm": 0.09428705275058746, | |
| "learning_rate": 4.343931245134616e-05, | |
| "loss": 1.0831, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.09291902184486389, | |
| "learning_rate": 4.028414082972141e-05, | |
| "loss": 1.1003, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7261904761904762, | |
| "grad_norm": 0.08907562494277954, | |
| "learning_rate": 3.721878753279017e-05, | |
| "loss": 1.0676, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.7380952380952381, | |
| "grad_norm": 0.09440898895263672, | |
| "learning_rate": 3.424786314309365e-05, | |
| "loss": 1.0129, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.09342656284570694, | |
| "learning_rate": 3.137583621312665e-05, | |
| "loss": 1.0515, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 0.09086596220731735, | |
| "learning_rate": 2.8607026544210114e-05, | |
| "loss": 1.0601, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.7738095238095238, | |
| "grad_norm": 0.09026821702718735, | |
| "learning_rate": 2.594559868909956e-05, | |
| "loss": 1.0162, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.7857142857142857, | |
| "grad_norm": 0.09036453813314438, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 1.0327, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7976190476190477, | |
| "grad_norm": 0.08887065201997757, | |
| "learning_rate": 2.0960733048124083e-05, | |
| "loss": 1.0744, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.8095238095238095, | |
| "grad_norm": 0.08880475908517838, | |
| "learning_rate": 1.864479297370325e-05, | |
| "loss": 1.0612, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.8095238095238095, | |
| "eval_loss": 1.031933069229126, | |
| "eval_runtime": 248.1305, | |
| "eval_samples_per_second": 2.708, | |
| "eval_steps_per_second": 0.339, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.8214285714285714, | |
| "grad_norm": 0.0960259959101677, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 1.0564, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.09149077534675598, | |
| "learning_rate": 1.4383310046973365e-05, | |
| "loss": 1.0129, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8452380952380952, | |
| "grad_norm": 0.08991867303848267, | |
| "learning_rate": 1.2444176869790925e-05, | |
| "loss": 1.0187, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 0.09646643698215485, | |
| "learning_rate": 1.0636735967658784e-05, | |
| "loss": 1.0561, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8690476190476191, | |
| "grad_norm": 0.09574442356824875, | |
| "learning_rate": 8.963705903385345e-06, | |
| "loss": 1.0474, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.8809523809523809, | |
| "grad_norm": 0.0916275605559349, | |
| "learning_rate": 7.427603073110967e-06, | |
| "loss": 1.0217, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 0.099373459815979, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 1.0831, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9047619047619048, | |
| "grad_norm": 0.0911068245768547, | |
| "learning_rate": 4.775211466158469e-06, | |
| "loss": 1.0488, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.9166666666666666, | |
| "grad_norm": 0.0971531942486763, | |
| "learning_rate": 3.662912138411967e-06, | |
| "loss": 1.0365, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.9285714285714286, | |
| "grad_norm": 0.0948888435959816, | |
| "learning_rate": 2.6955129420176196e-06, | |
| "loss": 1.0705, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.9404761904761905, | |
| "grad_norm": 0.09453336149454117, | |
| "learning_rate": 1.874468937261531e-06, | |
| "loss": 1.0429, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.10001447796821594, | |
| "learning_rate": 1.201015052319099e-06, | |
| "loss": 0.985, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9642857142857143, | |
| "grad_norm": 0.09593307226896286, | |
| "learning_rate": 6.761642258056978e-07, | |
| "loss": 1.0146, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9761904761904762, | |
| "grad_norm": 0.08949106186628342, | |
| "learning_rate": 3.007058832207976e-07, | |
| "loss": 1.0151, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.9880952380952381, | |
| "grad_norm": 0.09485086053609848, | |
| "learning_rate": 7.520474957699586e-08, | |
| "loss": 1.0602, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.0924575999379158, | |
| "learning_rate": 0.0, | |
| "loss": 1.0294, | |
| "step": 84 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 84, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 6.334928379474739e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |