| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09995002498750624, | |
| "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": 3.3340563149001086, | |
| "learning_rate": 0.0, | |
| "loss": 11.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.398812329952019, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 10.125, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 2.394322446895115, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 10.1172, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.9958816684399585, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 9.875, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 1.8270465897882062, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 9.6641, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.7854046471397795, | |
| "learning_rate": 0.0003, | |
| "loss": 9.4844, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.719416749115252, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 9.3281, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.4637825746112274, | |
| "learning_rate": 0.00041999999999999996, | |
| "loss": 9.2109, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.4393631015406718, | |
| "learning_rate": 0.00047999999999999996, | |
| "loss": 8.9453, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.2936734586915988, | |
| "learning_rate": 0.00054, | |
| "loss": 8.7109, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.0756922378227356, | |
| "learning_rate": 0.0005999986405514987, | |
| "loss": 8.4609, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9277829127413892, | |
| "learning_rate": 0.0005999945622196846, | |
| "loss": 8.2344, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8084581786682467, | |
| "learning_rate": 0.0005999877650456265, | |
| "loss": 8.125, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.7635084596900947, | |
| "learning_rate": 0.000599978249097772, | |
| "loss": 7.9766, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.9186699644247788, | |
| "learning_rate": 0.0005999660144719463, | |
| "loss": 7.8555, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6609504256551479, | |
| "learning_rate": 0.0005999510612913519, | |
| "loss": 7.7734, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7086232844782971, | |
| "learning_rate": 0.0005999333897065673, | |
| "loss": 7.7148, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 16.38048851691348, | |
| "learning_rate": 0.0005999129998955453, | |
| "loss": 8.4844, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.3057527590449889, | |
| "learning_rate": 0.0005998898920636111, | |
| "loss": 7.7539, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6966048242948986, | |
| "learning_rate": 0.00059986406644346, | |
| "loss": 7.75, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6348089115348993, | |
| "learning_rate": 0.0005998355232951559, | |
| "loss": 7.7031, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.7829163518610293, | |
| "learning_rate": 0.0005998042629061279, | |
| "loss": 7.6992, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.5900591778980369, | |
| "learning_rate": 0.0005997702855911678, | |
| "loss": 7.6016, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.4655170213064256, | |
| "learning_rate": 0.0005997335916924268, | |
| "loss": 7.5977, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.6287348258915756, | |
| "learning_rate": 0.0005996941815794121, | |
| "loss": 7.5586, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.6137321903884564, | |
| "learning_rate": 0.0005996520556489831, | |
| "loss": 7.5898, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.44962562710631065, | |
| "learning_rate": 0.0005996072143253473, | |
| "loss": 7.4336, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.46130046454703316, | |
| "learning_rate": 0.0005995596580600566, | |
| "loss": 7.4023, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4686712675731326, | |
| "learning_rate": 0.0005995093873320018, | |
| "loss": 7.3789, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4672147564288997, | |
| "learning_rate": 0.0005994564026474087, | |
| "loss": 7.3711, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.40408354581233474, | |
| "learning_rate": 0.0005994007045398324, | |
| "loss": 7.3672, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.46032146732584733, | |
| "learning_rate": 0.0005993422935701524, | |
| "loss": 7.3477, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4765534634593268, | |
| "learning_rate": 0.0005992811703265664, | |
| "loss": 7.3555, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.46208489386235113, | |
| "learning_rate": 0.0005992173354245849, | |
| "loss": 7.3047, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.2956144524964961, | |
| "learning_rate": 0.0005991507895070244, | |
| "loss": 7.3125, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.4834645389868856, | |
| "learning_rate": 0.0005990815332440017, | |
| "loss": 7.207, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.4411831350968505, | |
| "learning_rate": 0.0005990095673329266, | |
| "loss": 7.1758, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.24809297748968667, | |
| "learning_rate": 0.0005989348924984951, | |
| "loss": 7.2188, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.39402988416840584, | |
| "learning_rate": 0.0005988575094926817, | |
| "loss": 7.1953, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3868345222189167, | |
| "learning_rate": 0.0005987774190947328, | |
| "loss": 7.1641, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3777261230135448, | |
| "learning_rate": 0.0005986946221111575, | |
| "loss": 7.1328, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.4687511444077827, | |
| "learning_rate": 0.0005986091193757206, | |
| "loss": 7.0898, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.34935796211612463, | |
| "learning_rate": 0.0005985209117494337, | |
| "loss": 7.1367, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.38764476686849886, | |
| "learning_rate": 0.0005984300001205466, | |
| "loss": 7.125, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3956487898882936, | |
| "learning_rate": 0.0005983363854045386, | |
| "loss": 7.1094, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.31140257544677513, | |
| "learning_rate": 0.0005982400685441084, | |
| "loss": 7.0898, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.3664476570531787, | |
| "learning_rate": 0.0005981410505091662, | |
| "loss": 7.0664, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.31891741142945207, | |
| "learning_rate": 0.0005980393322968223, | |
| "loss": 7.0273, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4533529037337155, | |
| "learning_rate": 0.0005979349149313778, | |
| "loss": 7.0586, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.30532331638835586, | |
| "learning_rate": 0.0005978277994643147, | |
| "loss": 7.0195, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.6501991746260075, | |
| "learning_rate": 0.0005977179869742844, | |
| "loss": 6.9648, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.43904455901717926, | |
| "learning_rate": 0.0005976054785670975, | |
| "loss": 6.9805, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4826001598483571, | |
| "learning_rate": 0.0005974902753757124, | |
| "loss": 6.9297, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.2924998027034648, | |
| "learning_rate": 0.000597372378560224, | |
| "loss": 6.8984, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4439033666380787, | |
| "learning_rate": 0.0005972517893078517, | |
| "loss": 6.8945, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.6135914255073411, | |
| "learning_rate": 0.0005971285088329284, | |
| "loss": 6.9727, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5575686565598483, | |
| "learning_rate": 0.0005970025383768866, | |
| "loss": 6.9219, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.4820951675994578, | |
| "learning_rate": 0.0005968738792082478, | |
| "loss": 6.8516, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.40164190019465584, | |
| "learning_rate": 0.0005967425326226082, | |
| "loss": 6.7734, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.46129863945181293, | |
| "learning_rate": 0.0005966084999426265, | |
| "loss": 6.8125, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.33322355827118677, | |
| "learning_rate": 0.0005964717825180101, | |
| "loss": 6.7891, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.3847525153855558, | |
| "learning_rate": 0.0005963323817255024, | |
| "loss": 6.8242, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.3384433591375982, | |
| "learning_rate": 0.0005961902989688674, | |
| "loss": 6.707, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.3937003195165685, | |
| "learning_rate": 0.000596045535678877, | |
| "loss": 6.8203, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.35423488053528107, | |
| "learning_rate": 0.0005958980933132962, | |
| "loss": 6.7383, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.36005939745315396, | |
| "learning_rate": 0.0005957479733568675, | |
| "loss": 6.7109, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3499278317706933, | |
| "learning_rate": 0.0005955951773212976, | |
| "loss": 6.7266, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3708385192137018, | |
| "learning_rate": 0.0005954397067452407, | |
| "loss": 6.7617, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3775657656205869, | |
| "learning_rate": 0.0005952815631942839, | |
| "loss": 6.7148, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3040083750375816, | |
| "learning_rate": 0.0005951207482609307, | |
| "loss": 6.5938, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3443020808841468, | |
| "learning_rate": 0.0005949572635645861, | |
| "loss": 6.6523, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3520066316939, | |
| "learning_rate": 0.0005947911107515389, | |
| "loss": 6.6211, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3739040572679613, | |
| "learning_rate": 0.0005946222914949462, | |
| "loss": 6.5547, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.34890731989025553, | |
| "learning_rate": 0.000594450807494816, | |
| "loss": 6.5859, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.40910932350136514, | |
| "learning_rate": 0.0005942766604779903, | |
| "loss": 6.5547, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5698342865852906, | |
| "learning_rate": 0.0005940998521981274, | |
| "loss": 6.457, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5179452709555474, | |
| "learning_rate": 0.0005939203844356852, | |
| "loss": 6.5547, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5222512938673792, | |
| "learning_rate": 0.0005937382589979016, | |
| "loss": 6.5039, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.5682332793686307, | |
| "learning_rate": 0.0005935534777187781, | |
| "loss": 6.5547, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3869287710460676, | |
| "learning_rate": 0.0005933660424590598, | |
| "loss": 6.5156, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3078211032807607, | |
| "learning_rate": 0.000593175955106218, | |
| "loss": 6.4258, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3611357511872241, | |
| "learning_rate": 0.00059298321757443, | |
| "loss": 6.4727, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.29633467844266953, | |
| "learning_rate": 0.0005927878318045608, | |
| "loss": 6.3281, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3257574200776832, | |
| "learning_rate": 0.0005925897997641426, | |
| "loss": 6.3203, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.2824054533852328, | |
| "learning_rate": 0.0005923891234473562, | |
| "loss": 6.4062, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.3056199770204573, | |
| "learning_rate": 0.0005921858048750097, | |
| "loss": 6.3984, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.2966438824341908, | |
| "learning_rate": 0.000591979846094519, | |
| "loss": 6.3555, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.32782438676663733, | |
| "learning_rate": 0.0005917712491798866, | |
| "loss": 6.4023, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.3538316399620157, | |
| "learning_rate": 0.0005915600162316811, | |
| "loss": 6.2812, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.375858298192913, | |
| "learning_rate": 0.0005913461493770162, | |
| "loss": 6.3086, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.5189251339815161, | |
| "learning_rate": 0.0005911296507695284, | |
| "loss": 6.2812, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.6304909542669104, | |
| "learning_rate": 0.0005909105225893564, | |
| "loss": 6.2969, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.4655662819622591, | |
| "learning_rate": 0.0005906887670431187, | |
| "loss": 6.1953, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.39035390983920965, | |
| "learning_rate": 0.000590464386363891, | |
| "loss": 6.2617, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.4918417851770978, | |
| "learning_rate": 0.0005902373828111843, | |
| "loss": 6.2148, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.35670770889552555, | |
| "learning_rate": 0.0005900077586709219, | |
| "loss": 6.2461, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.4177985869939347, | |
| "learning_rate": 0.0005897755162554163, | |
| "loss": 6.1797, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.3742471130708234, | |
| "learning_rate": 0.000589540657903346, | |
| "loss": 6.1406, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.28627666723978284, | |
| "learning_rate": 0.0005893031859797322, | |
| "loss": 6.2031, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.32238563846046103, | |
| "learning_rate": 0.0005890631028759143, | |
| "loss": 6.0625, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |