| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.019436345966958212, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.820549726486206, | |
| "learning_rate": 0.00019805941782534764, | |
| "loss": 4.7714, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.7905406951904297, | |
| "learning_rate": 0.00019605881764529358, | |
| "loss": 4.8623, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.4746589660644531, | |
| "learning_rate": 0.00019405821746523957, | |
| "loss": 3.7993, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.333172559738159, | |
| "learning_rate": 0.00019205761728518557, | |
| "loss": 4.5898, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.711241602897644, | |
| "learning_rate": 0.00019005701710513156, | |
| "loss": 3.1208, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.001378059387207, | |
| "learning_rate": 0.00018805641692507753, | |
| "loss": 2.7962, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.5922505855560303, | |
| "learning_rate": 0.00018605581674502352, | |
| "loss": 2.7348, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 3.073798179626465, | |
| "learning_rate": 0.00018405521656496952, | |
| "loss": 2.4213, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.494370937347412, | |
| "learning_rate": 0.00018205461638491548, | |
| "loss": 2.2847, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 3.1938252449035645, | |
| "learning_rate": 0.00018005401620486148, | |
| "loss": 2.2266, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.0904405117034912, | |
| "learning_rate": 0.00017805341602480744, | |
| "loss": 1.7216, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8593369126319885, | |
| "learning_rate": 0.00017605281584475344, | |
| "loss": 2.0195, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.051988959312439, | |
| "learning_rate": 0.00017405221566469943, | |
| "loss": 1.8962, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.313820242881775, | |
| "learning_rate": 0.00017205161548464542, | |
| "loss": 1.735, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.035528302192688, | |
| "learning_rate": 0.00017005101530459136, | |
| "loss": 2.0408, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8312838077545166, | |
| "learning_rate": 0.00016805041512453736, | |
| "loss": 1.7348, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.406984806060791, | |
| "learning_rate": 0.00016604981494448335, | |
| "loss": 1.9893, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.0014063119888306, | |
| "learning_rate": 0.00016404921476442935, | |
| "loss": 1.7413, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.1123204231262207, | |
| "learning_rate": 0.0001620486145843753, | |
| "loss": 1.642, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.9622805118560791, | |
| "learning_rate": 0.0001600480144043213, | |
| "loss": 1.8947, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.9021405577659607, | |
| "learning_rate": 0.0001580474142242673, | |
| "loss": 1.8771, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.7586336135864258, | |
| "learning_rate": 0.00015604681404421327, | |
| "loss": 1.7902, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8033269643783569, | |
| "learning_rate": 0.00015404621386415926, | |
| "loss": 1.536, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.8731617331504822, | |
| "learning_rate": 0.00015204561368410523, | |
| "loss": 1.6825, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.7662222981452942, | |
| "learning_rate": 0.00015004501350405122, | |
| "loss": 1.6269, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8472873568534851, | |
| "learning_rate": 0.00014804441332399721, | |
| "loss": 1.7143, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8353914618492126, | |
| "learning_rate": 0.0001460438131439432, | |
| "loss": 0.9914, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8981612324714661, | |
| "learning_rate": 0.00014404321296388918, | |
| "loss": 1.1645, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7975760698318481, | |
| "learning_rate": 0.00014204261278383514, | |
| "loss": 1.4233, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7593973875045776, | |
| "learning_rate": 0.00014004201260378114, | |
| "loss": 1.4445, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9403287172317505, | |
| "learning_rate": 0.00013804141242372713, | |
| "loss": 1.4302, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9004844427108765, | |
| "learning_rate": 0.00013604081224367312, | |
| "loss": 1.5638, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8177943229675293, | |
| "learning_rate": 0.0001340402120636191, | |
| "loss": 1.3372, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.733913004398346, | |
| "learning_rate": 0.00013203961188356508, | |
| "loss": 1.2174, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7267743945121765, | |
| "learning_rate": 0.00013003901170351108, | |
| "loss": 1.0346, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9789184331893921, | |
| "learning_rate": 0.00012803841152345704, | |
| "loss": 1.4935, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7921544909477234, | |
| "learning_rate": 0.000126037811343403, | |
| "loss": 1.2153, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.687468945980072, | |
| "learning_rate": 0.000124037211163349, | |
| "loss": 1.096, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7953901290893555, | |
| "learning_rate": 0.000122036610983295, | |
| "loss": 1.3469, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6782249212265015, | |
| "learning_rate": 0.00012003601080324098, | |
| "loss": 1.1761, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7382201552391052, | |
| "learning_rate": 0.00011803541062318697, | |
| "loss": 1.2891, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7965229749679565, | |
| "learning_rate": 0.00011603481044313295, | |
| "loss": 1.2347, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6263877153396606, | |
| "learning_rate": 0.00011403421026307892, | |
| "loss": 1.2815, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8494048714637756, | |
| "learning_rate": 0.00011203361008302491, | |
| "loss": 1.4406, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8479952216148376, | |
| "learning_rate": 0.0001100330099029709, | |
| "loss": 1.1149, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6989169716835022, | |
| "learning_rate": 0.00010803240972291689, | |
| "loss": 1.3057, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8104137182235718, | |
| "learning_rate": 0.00010603180954286287, | |
| "loss": 1.1267, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8028876781463623, | |
| "learning_rate": 0.00010403120936280886, | |
| "loss": 1.6476, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7756268382072449, | |
| "learning_rate": 0.00010203060918275482, | |
| "loss": 1.2497, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7349205613136292, | |
| "learning_rate": 0.00010003000900270081, | |
| "loss": 0.9531, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6500608325004578, | |
| "learning_rate": 9.802940882264679e-05, | |
| "loss": 1.026, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6990982294082642, | |
| "learning_rate": 9.602880864259278e-05, | |
| "loss": 1.119, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.040234923362732, | |
| "learning_rate": 9.402820846253876e-05, | |
| "loss": 1.9937, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6553092002868652, | |
| "learning_rate": 9.202760828248476e-05, | |
| "loss": 1.1563, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7430468797683716, | |
| "learning_rate": 9.002700810243074e-05, | |
| "loss": 1.0199, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.748374342918396, | |
| "learning_rate": 8.802640792237672e-05, | |
| "loss": 1.2218, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6465599536895752, | |
| "learning_rate": 8.602580774232271e-05, | |
| "loss": 1.3286, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8411876559257507, | |
| "learning_rate": 8.402520756226868e-05, | |
| "loss": 1.2057, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8302045464515686, | |
| "learning_rate": 8.202460738221467e-05, | |
| "loss": 1.5277, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7100818753242493, | |
| "learning_rate": 8.002400720216065e-05, | |
| "loss": 1.3326, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6247999668121338, | |
| "learning_rate": 7.802340702210663e-05, | |
| "loss": 1.0997, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.1579809188842773, | |
| "learning_rate": 7.602280684205261e-05, | |
| "loss": 1.565, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7506289482116699, | |
| "learning_rate": 7.402220666199861e-05, | |
| "loss": 1.2788, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7227886915206909, | |
| "learning_rate": 7.202160648194459e-05, | |
| "loss": 0.9065, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.144627332687378, | |
| "learning_rate": 7.002100630189057e-05, | |
| "loss": 1.6695, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7828360199928284, | |
| "learning_rate": 6.802040612183656e-05, | |
| "loss": 1.0203, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.2042758464813232, | |
| "learning_rate": 6.601980594178254e-05, | |
| "loss": 1.2251, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6878300905227661, | |
| "learning_rate": 6.401920576172852e-05, | |
| "loss": 0.8927, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7350767850875854, | |
| "learning_rate": 6.20186055816745e-05, | |
| "loss": 1.0885, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7477673292160034, | |
| "learning_rate": 6.001800540162049e-05, | |
| "loss": 1.2264, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8758525848388672, | |
| "learning_rate": 5.801740522156648e-05, | |
| "loss": 1.3656, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.742152214050293, | |
| "learning_rate": 5.601680504151246e-05, | |
| "loss": 1.1293, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7166507840156555, | |
| "learning_rate": 5.4016204861458444e-05, | |
| "loss": 1.2553, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6855165362358093, | |
| "learning_rate": 5.201560468140443e-05, | |
| "loss": 1.2659, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8550592064857483, | |
| "learning_rate": 5.0015004501350405e-05, | |
| "loss": 1.2631, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.6638922691345215, | |
| "learning_rate": 4.801440432129639e-05, | |
| "loss": 0.972, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7261601090431213, | |
| "learning_rate": 4.601380414124238e-05, | |
| "loss": 1.2276, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7439085245132446, | |
| "learning_rate": 4.401320396118836e-05, | |
| "loss": 1.0153, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6688580513000488, | |
| "learning_rate": 4.201260378113434e-05, | |
| "loss": 0.9473, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.1068280935287476, | |
| "learning_rate": 4.0012003601080326e-05, | |
| "loss": 1.3205, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6699922680854797, | |
| "learning_rate": 3.801140342102631e-05, | |
| "loss": 1.2403, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.8933495879173279, | |
| "learning_rate": 3.6010803240972294e-05, | |
| "loss": 1.0529, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7752862572669983, | |
| "learning_rate": 3.401020306091828e-05, | |
| "loss": 1.1408, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.043106198310852, | |
| "learning_rate": 3.200960288086426e-05, | |
| "loss": 1.0511, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6093169450759888, | |
| "learning_rate": 3.0009002700810245e-05, | |
| "loss": 0.9798, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.9623333811759949, | |
| "learning_rate": 2.800840252075623e-05, | |
| "loss": 0.8907, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7118967175483704, | |
| "learning_rate": 2.6007802340702216e-05, | |
| "loss": 1.224, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7716160416603088, | |
| "learning_rate": 2.4007202160648196e-05, | |
| "loss": 1.2367, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.8782191872596741, | |
| "learning_rate": 2.200660198059418e-05, | |
| "loss": 1.2046, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.8954662084579468, | |
| "learning_rate": 2.0006001800540163e-05, | |
| "loss": 1.1141, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7129700183868408, | |
| "learning_rate": 1.8005401620486147e-05, | |
| "loss": 1.2983, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.8731375932693481, | |
| "learning_rate": 1.600480144043213e-05, | |
| "loss": 1.1551, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6418743133544922, | |
| "learning_rate": 1.4004201260378114e-05, | |
| "loss": 1.0709, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.5780905485153198, | |
| "learning_rate": 1.2003601080324098e-05, | |
| "loss": 0.7904, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6729098558425903, | |
| "learning_rate": 1.0003000900270082e-05, | |
| "loss": 1.2541, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.9264962673187256, | |
| "learning_rate": 8.002400720216065e-06, | |
| "loss": 0.9846, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6398671865463257, | |
| "learning_rate": 6.001800540162049e-06, | |
| "loss": 1.007, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.639799952507019, | |
| "learning_rate": 4.001200360108033e-06, | |
| "loss": 1.0604, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6854999661445618, | |
| "learning_rate": 2.0006001800540163e-06, | |
| "loss": 0.8762, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7004562020301819, | |
| "learning_rate": 0.0, | |
| "loss": 0.9948, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 1.5977891623471104e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |