| { |
| "best_global_step": 110, |
| "best_metric": 0.0009438548004254699, |
| "best_model_checkpoint": "./lora_adapter/checkpoint-110", |
| "epoch": 3.8177777777777777, |
| "eval_steps": 10, |
| "global_step": 110, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0, |
| "eval_loss": 5.712841987609863, |
| "eval_runtime": 5.5217, |
| "eval_samples_per_second": 9.055, |
| "eval_steps_per_second": 9.055, |
| "step": 0 |
| }, |
| { |
| "epoch": 0, |
| "eval_perplexity": 302.7302023645728, |
| "step": 0 |
| }, |
| { |
| "epoch": 0.035555555555555556, |
| "grad_norm": 0.3103635609149933, |
| "learning_rate": 0.0, |
| "loss": 5.682491779327393, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.07111111111111111, |
| "grad_norm": 0.4182112216949463, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 6.0904059410095215, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.10666666666666667, |
| "grad_norm": 0.403287798166275, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 5.459848403930664, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.14222222222222222, |
| "grad_norm": 0.4156350791454315, |
| "learning_rate": 2.5e-05, |
| "loss": 5.875328063964844, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.17777777777777778, |
| "grad_norm": 0.35492172837257385, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 5.3188605308532715, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.21333333333333335, |
| "grad_norm": 0.3586452007293701, |
| "learning_rate": 4.166666666666667e-05, |
| "loss": 5.076211929321289, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.24888888888888888, |
| "grad_norm": 0.40670934319496155, |
| "learning_rate": 5e-05, |
| "loss": 4.494481563568115, |
| "step": 7 |
| }, |
| { |
| "epoch": 0.28444444444444444, |
| "grad_norm": 0.4478435218334198, |
| "learning_rate": 5.833333333333334e-05, |
| "loss": 4.102505207061768, |
| "step": 8 |
| }, |
| { |
| "epoch": 0.32, |
| "grad_norm": 0.4579038619995117, |
| "learning_rate": 6.666666666666667e-05, |
| "loss": 3.379936695098877, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.35555555555555557, |
| "grad_norm": 0.38791775703430176, |
| "learning_rate": 7.500000000000001e-05, |
| "loss": 2.952427387237549, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.35555555555555557, |
| "eval_loss": 2.718632221221924, |
| "eval_runtime": 5.4982, |
| "eval_samples_per_second": 9.094, |
| "eval_steps_per_second": 9.094, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.35555555555555557, |
| "eval_perplexity": 15.15957311568705, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.39111111111111113, |
| "grad_norm": 0.7142107486724854, |
| "learning_rate": 8.333333333333334e-05, |
| "loss": 2.6527280807495117, |
| "step": 11 |
| }, |
| { |
| "epoch": 0.4266666666666667, |
| "grad_norm": 0.46958744525909424, |
| "learning_rate": 9.166666666666667e-05, |
| "loss": 2.0736618041992188, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.4622222222222222, |
| "grad_norm": 0.46414491534233093, |
| "learning_rate": 0.0001, |
| "loss": 1.5400762557983398, |
| "step": 13 |
| }, |
| { |
| "epoch": 0.49777777777777776, |
| "grad_norm": 0.3608834445476532, |
| "learning_rate": 9.999490215047167e-05, |
| "loss": 1.2599258422851562, |
| "step": 14 |
| }, |
| { |
| "epoch": 0.5333333333333333, |
| "grad_norm": 0.3089125156402588, |
| "learning_rate": 9.997960964140947e-05, |
| "loss": 1.0378409624099731, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.5688888888888889, |
| "grad_norm": 0.23846635222434998, |
| "learning_rate": 9.995412559116979e-05, |
| "loss": 0.683804452419281, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.6044444444444445, |
| "grad_norm": 0.25005361437797546, |
| "learning_rate": 9.991845519630678e-05, |
| "loss": 0.6251506805419922, |
| "step": 17 |
| }, |
| { |
| "epoch": 0.64, |
| "grad_norm": 0.17282210290431976, |
| "learning_rate": 9.987260573051269e-05, |
| "loss": 0.40306615829467773, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.6755555555555556, |
| "grad_norm": 0.14316385984420776, |
| "learning_rate": 9.981658654313457e-05, |
| "loss": 0.36748170852661133, |
| "step": 19 |
| }, |
| { |
| "epoch": 0.7111111111111111, |
| "grad_norm": 0.09027505666017532, |
| "learning_rate": 9.975040905726798e-05, |
| "loss": 0.3596840500831604, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.7111111111111111, |
| "eval_loss": 0.34836798906326294, |
| "eval_runtime": 5.5169, |
| "eval_samples_per_second": 9.063, |
| "eval_steps_per_second": 9.063, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.7111111111111111, |
| "eval_perplexity": 1.4167535036209495, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.7466666666666667, |
| "grad_norm": 0.07985851913690567, |
| "learning_rate": 9.967408676742751e-05, |
| "loss": 0.36926567554473877, |
| "step": 21 |
| }, |
| { |
| "epoch": 0.7822222222222223, |
| "grad_norm": 0.06538072973489761, |
| "learning_rate": 9.958763523679514e-05, |
| "loss": 0.3178553879261017, |
| "step": 22 |
| }, |
| { |
| "epoch": 0.8177777777777778, |
| "grad_norm": 0.05077500641345978, |
| "learning_rate": 9.949107209404665e-05, |
| "loss": 0.19684894382953644, |
| "step": 23 |
| }, |
| { |
| "epoch": 0.8533333333333334, |
| "grad_norm": 0.10189707577228546, |
| "learning_rate": 9.938441702975689e-05, |
| "loss": 0.2359190434217453, |
| "step": 24 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 0.06018666550517082, |
| "learning_rate": 9.926769179238466e-05, |
| "loss": 0.24584323167800903, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.9244444444444444, |
| "grad_norm": 0.05413977429270744, |
| "learning_rate": 9.914092018383778e-05, |
| "loss": 0.17237712442874908, |
| "step": 26 |
| }, |
| { |
| "epoch": 0.96, |
| "grad_norm": 0.042994916439056396, |
| "learning_rate": 9.900412805461967e-05, |
| "loss": 0.15629145503044128, |
| "step": 27 |
| }, |
| { |
| "epoch": 0.9955555555555555, |
| "grad_norm": 0.0588720478117466, |
| "learning_rate": 9.885734329855798e-05, |
| "loss": 0.1863369643688202, |
| "step": 28 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.15661829710006714, |
| "learning_rate": 9.870059584711668e-05, |
| "loss": 0.2297505885362625, |
| "step": 29 |
| }, |
| { |
| "epoch": 1.0355555555555556, |
| "grad_norm": 0.03409217298030853, |
| "learning_rate": 9.853391766329263e-05, |
| "loss": 0.06839371472597122, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.0355555555555556, |
| "eval_loss": 0.0960446149110794, |
| "eval_runtime": 5.4537, |
| "eval_samples_per_second": 9.168, |
| "eval_steps_per_second": 9.168, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.0355555555555556, |
| "eval_perplexity": 1.1008081753572803, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.0711111111111111, |
| "grad_norm": 0.03011409193277359, |
| "learning_rate": 9.835734273509786e-05, |
| "loss": 0.07586748898029327, |
| "step": 31 |
| }, |
| { |
| "epoch": 1.1066666666666667, |
| "grad_norm": 0.03701885789632797, |
| "learning_rate": 9.817090706862895e-05, |
| "loss": 0.0892762839794159, |
| "step": 32 |
| }, |
| { |
| "epoch": 1.1422222222222222, |
| "grad_norm": 0.03201104328036308, |
| "learning_rate": 9.797464868072488e-05, |
| "loss": 0.08438154309988022, |
| "step": 33 |
| }, |
| { |
| "epoch": 1.1777777777777778, |
| "grad_norm": 0.03157950937747955, |
| "learning_rate": 9.776860759121484e-05, |
| "loss": 0.04694896563887596, |
| "step": 34 |
| }, |
| { |
| "epoch": 1.2133333333333334, |
| "grad_norm": 0.032276663929224014, |
| "learning_rate": 9.755282581475769e-05, |
| "loss": 0.07806245237588882, |
| "step": 35 |
| }, |
| { |
| "epoch": 1.248888888888889, |
| "grad_norm": 0.030301420018076897, |
| "learning_rate": 9.73273473522745e-05, |
| "loss": 0.0493946447968483, |
| "step": 36 |
| }, |
| { |
| "epoch": 1.2844444444444445, |
| "grad_norm": 0.027570156380534172, |
| "learning_rate": 9.709221818197624e-05, |
| "loss": 0.06280327588319778, |
| "step": 37 |
| }, |
| { |
| "epoch": 1.32, |
| "grad_norm": 0.025066696107387543, |
| "learning_rate": 9.68474862499881e-05, |
| "loss": 0.05094042792916298, |
| "step": 38 |
| }, |
| { |
| "epoch": 1.3555555555555556, |
| "grad_norm": 0.02456921711564064, |
| "learning_rate": 9.659320146057262e-05, |
| "loss": 0.03677443787455559, |
| "step": 39 |
| }, |
| { |
| "epoch": 1.3911111111111112, |
| "grad_norm": 0.02479228563606739, |
| "learning_rate": 9.632941566595357e-05, |
| "loss": 0.03535553067922592, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.3911111111111112, |
| "eval_loss": 0.022668220102787018, |
| "eval_runtime": 5.6373, |
| "eval_samples_per_second": 8.87, |
| "eval_steps_per_second": 8.87, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.3911111111111112, |
| "eval_perplexity": 1.022927096593193, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.4266666666666667, |
| "grad_norm": 0.016291167587041855, |
| "learning_rate": 9.60561826557425e-05, |
| "loss": 0.019891822710633278, |
| "step": 41 |
| }, |
| { |
| "epoch": 1.462222222222222, |
| "grad_norm": 0.01630846969783306, |
| "learning_rate": 9.577355814597031e-05, |
| "loss": 0.023395322263240814, |
| "step": 42 |
| }, |
| { |
| "epoch": 1.4977777777777779, |
| "grad_norm": 0.014682851731777191, |
| "learning_rate": 9.548159976772592e-05, |
| "loss": 0.01620781607925892, |
| "step": 43 |
| }, |
| { |
| "epoch": 1.5333333333333332, |
| "grad_norm": 0.004914387129247189, |
| "learning_rate": 9.518036705540458e-05, |
| "loss": 0.006628153380006552, |
| "step": 44 |
| }, |
| { |
| "epoch": 1.568888888888889, |
| "grad_norm": 0.005044769961386919, |
| "learning_rate": 9.486992143456792e-05, |
| "loss": 0.0069605098105967045, |
| "step": 45 |
| }, |
| { |
| "epoch": 1.6044444444444443, |
| "grad_norm": 0.010346329770982265, |
| "learning_rate": 9.45503262094184e-05, |
| "loss": 0.009333918802440166, |
| "step": 46 |
| }, |
| { |
| "epoch": 1.6400000000000001, |
| "grad_norm": 0.005329441279172897, |
| "learning_rate": 9.422164654989072e-05, |
| "loss": 0.006601989734917879, |
| "step": 47 |
| }, |
| { |
| "epoch": 1.6755555555555555, |
| "grad_norm": 0.008117754012346268, |
| "learning_rate": 9.388394947836279e-05, |
| "loss": 0.006798232439905405, |
| "step": 48 |
| }, |
| { |
| "epoch": 1.7111111111111112, |
| "grad_norm": 0.008304511196911335, |
| "learning_rate": 9.353730385598887e-05, |
| "loss": 0.006939806509763002, |
| "step": 49 |
| }, |
| { |
| "epoch": 1.7466666666666666, |
| "grad_norm": 0.005011966452002525, |
| "learning_rate": 9.318178036865785e-05, |
| "loss": 0.005817799363285303, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.7466666666666666, |
| "eval_loss": 0.005097925662994385, |
| "eval_runtime": 5.4641, |
| "eval_samples_per_second": 9.151, |
| "eval_steps_per_second": 9.151, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.7466666666666666, |
| "eval_perplexity": 1.0051109421957325, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.7822222222222224, |
| "grad_norm": 0.00494053028523922, |
| "learning_rate": 9.281745151257946e-05, |
| "loss": 0.005493980832397938, |
| "step": 51 |
| }, |
| { |
| "epoch": 1.8177777777777777, |
| "grad_norm": 0.003944819793105125, |
| "learning_rate": 9.244439157950114e-05, |
| "loss": 0.004002867732197046, |
| "step": 52 |
| }, |
| { |
| "epoch": 1.8533333333333335, |
| "grad_norm": 0.002889649011194706, |
| "learning_rate": 9.206267664155907e-05, |
| "loss": 0.00337179284542799, |
| "step": 53 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 0.0023699572775512934, |
| "learning_rate": 9.167238453576589e-05, |
| "loss": 0.003095966763794422, |
| "step": 54 |
| }, |
| { |
| "epoch": 1.9244444444444444, |
| "grad_norm": 0.009545288980007172, |
| "learning_rate": 9.12735948481387e-05, |
| "loss": 0.005139315500855446, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.96, |
| "grad_norm": 0.0045837461948394775, |
| "learning_rate": 9.086638889747035e-05, |
| "loss": 0.0036253915168344975, |
| "step": 56 |
| }, |
| { |
| "epoch": 1.9955555555555555, |
| "grad_norm": 0.008223620243370533, |
| "learning_rate": 9.045084971874738e-05, |
| "loss": 0.004782550968229771, |
| "step": 57 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.005293046589940786, |
| "learning_rate": 9.002706204621803e-05, |
| "loss": 0.003609852399677038, |
| "step": 58 |
| }, |
| { |
| "epoch": 2.0355555555555553, |
| "grad_norm": 0.0016866261139512062, |
| "learning_rate": 8.959511229611376e-05, |
| "loss": 0.0023398897610604763, |
| "step": 59 |
| }, |
| { |
| "epoch": 2.071111111111111, |
| "grad_norm": 0.0024739305954426527, |
| "learning_rate": 8.915508854902778e-05, |
| "loss": 0.0026166446041315794, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.071111111111111, |
| "eval_loss": 0.0024453848600387573, |
| "eval_runtime": 5.4775, |
| "eval_samples_per_second": 9.128, |
| "eval_steps_per_second": 9.128, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.071111111111111, |
| "eval_perplexity": 1.002448377252282, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.1066666666666665, |
| "grad_norm": 0.0023085270076990128, |
| "learning_rate": 8.870708053195413e-05, |
| "loss": 0.002390390494838357, |
| "step": 61 |
| }, |
| { |
| "epoch": 2.1422222222222222, |
| "grad_norm": 0.0026797724422067404, |
| "learning_rate": 8.825117959999116e-05, |
| "loss": 0.002457966562360525, |
| "step": 62 |
| }, |
| { |
| "epoch": 2.1777777777777776, |
| "grad_norm": 0.0013235695660114288, |
| "learning_rate": 8.778747871771292e-05, |
| "loss": 0.0019208687590435147, |
| "step": 63 |
| }, |
| { |
| "epoch": 2.2133333333333334, |
| "grad_norm": 0.0019502972718328238, |
| "learning_rate": 8.731607244021236e-05, |
| "loss": 0.0020442644599825144, |
| "step": 64 |
| }, |
| { |
| "epoch": 2.2488888888888887, |
| "grad_norm": 0.0014056526124477386, |
| "learning_rate": 8.683705689382024e-05, |
| "loss": 0.0020194968674331903, |
| "step": 65 |
| }, |
| { |
| "epoch": 2.2844444444444445, |
| "grad_norm": 0.0016124699031934142, |
| "learning_rate": 8.635052975650369e-05, |
| "loss": 0.001853243331424892, |
| "step": 66 |
| }, |
| { |
| "epoch": 2.32, |
| "grad_norm": 0.0014297482557594776, |
| "learning_rate": 8.585659023794818e-05, |
| "loss": 0.002052597003057599, |
| "step": 67 |
| }, |
| { |
| "epoch": 2.3555555555555556, |
| "grad_norm": 0.0011592477094382048, |
| "learning_rate": 8.535533905932738e-05, |
| "loss": 0.0016631247708573937, |
| "step": 68 |
| }, |
| { |
| "epoch": 2.391111111111111, |
| "grad_norm": 0.0014637873973697424, |
| "learning_rate": 8.484687843276469e-05, |
| "loss": 0.0018859267001971602, |
| "step": 69 |
| }, |
| { |
| "epoch": 2.4266666666666667, |
| "grad_norm": 0.0009684404940344393, |
| "learning_rate": 8.433131204049067e-05, |
| "loss": 0.001511464361101389, |
| "step": 70 |
| }, |
| { |
| "epoch": 2.4266666666666667, |
| "eval_loss": 0.0016330081271007657, |
| "eval_runtime": 5.4569, |
| "eval_samples_per_second": 9.163, |
| "eval_steps_per_second": 9.163, |
| "step": 70 |
| }, |
| { |
| "epoch": 2.4266666666666667, |
| "eval_perplexity": 1.0016343422109635, |
| "step": 70 |
| }, |
| { |
| "epoch": 2.462222222222222, |
| "grad_norm": 0.0010153243783861399, |
| "learning_rate": 8.380874501370097e-05, |
| "loss": 0.0015972007531672716, |
| "step": 71 |
| }, |
| { |
| "epoch": 2.497777777777778, |
| "grad_norm": 0.0009011203073896468, |
| "learning_rate": 8.327928391111841e-05, |
| "loss": 0.0013946370454505086, |
| "step": 72 |
| }, |
| { |
| "epoch": 2.533333333333333, |
| "grad_norm": 0.0009017665288411081, |
| "learning_rate": 8.274303669726426e-05, |
| "loss": 0.0014331797137856483, |
| "step": 73 |
| }, |
| { |
| "epoch": 2.568888888888889, |
| "grad_norm": 0.0009415379608981311, |
| "learning_rate": 8.220011272044277e-05, |
| "loss": 0.001520266872830689, |
| "step": 74 |
| }, |
| { |
| "epoch": 2.6044444444444443, |
| "grad_norm": 0.0009506358182989061, |
| "learning_rate": 8.165062269044353e-05, |
| "loss": 0.001505439169704914, |
| "step": 75 |
| }, |
| { |
| "epoch": 2.64, |
| "grad_norm": 0.0007934041786938906, |
| "learning_rate": 8.109467865596612e-05, |
| "loss": 0.0013778579887002707, |
| "step": 76 |
| }, |
| { |
| "epoch": 2.6755555555555555, |
| "grad_norm": 0.0008148634806275368, |
| "learning_rate": 8.053239398177191e-05, |
| "loss": 0.001370853278785944, |
| "step": 77 |
| }, |
| { |
| "epoch": 2.7111111111111112, |
| "grad_norm": 0.0006598148611374199, |
| "learning_rate": 7.996388332556735e-05, |
| "loss": 0.0011620635632425547, |
| "step": 78 |
| }, |
| { |
| "epoch": 2.7466666666666666, |
| "grad_norm": 0.0008030621684156358, |
| "learning_rate": 7.938926261462366e-05, |
| "loss": 0.0012577238958328962, |
| "step": 79 |
| }, |
| { |
| "epoch": 2.7822222222222224, |
| "grad_norm": 0.0008109168848022819, |
| "learning_rate": 7.880864902213765e-05, |
| "loss": 0.0013199358945712447, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.7822222222222224, |
| "eval_loss": 0.0013468421529978514, |
| "eval_runtime": 5.4617, |
| "eval_samples_per_second": 9.155, |
| "eval_steps_per_second": 9.155, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.7822222222222224, |
| "eval_perplexity": 1.0013477495522192, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.8177777777777777, |
| "grad_norm": 0.0007556782802566886, |
| "learning_rate": 7.822216094333847e-05, |
| "loss": 0.001324485638178885, |
| "step": 81 |
| }, |
| { |
| "epoch": 2.8533333333333335, |
| "grad_norm": 0.0008008314180187881, |
| "learning_rate": 7.762991797134514e-05, |
| "loss": 0.0012606465024873614, |
| "step": 82 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.0007433093851432204, |
| "learning_rate": 7.703204087277988e-05, |
| "loss": 0.0012966024223715067, |
| "step": 83 |
| }, |
| { |
| "epoch": 2.924444444444444, |
| "grad_norm": 0.0007187623996287584, |
| "learning_rate": 7.64286515631421e-05, |
| "loss": 0.001278804033063352, |
| "step": 84 |
| }, |
| { |
| "epoch": 2.96, |
| "grad_norm": 0.0007242705905809999, |
| "learning_rate": 7.58198730819481e-05, |
| "loss": 0.0012370930053293705, |
| "step": 85 |
| }, |
| { |
| "epoch": 2.9955555555555557, |
| "grad_norm": 0.0005744029767811298, |
| "learning_rate": 7.52058295676416e-05, |
| "loss": 0.0010438471799716353, |
| "step": 86 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.0008678279700689018, |
| "learning_rate": 7.45866462322802e-05, |
| "loss": 0.0009476285194978118, |
| "step": 87 |
| }, |
| { |
| "epoch": 3.0355555555555553, |
| "grad_norm": 0.0006672442541457713, |
| "learning_rate": 7.396244933600285e-05, |
| "loss": 0.0011832310119643807, |
| "step": 88 |
| }, |
| { |
| "epoch": 3.071111111111111, |
| "grad_norm": 0.0007027190877124667, |
| "learning_rate": 7.333336616128369e-05, |
| "loss": 0.0011928146705031395, |
| "step": 89 |
| }, |
| { |
| "epoch": 3.1066666666666665, |
| "grad_norm": 0.0005659974412992597, |
| "learning_rate": 7.269952498697734e-05, |
| "loss": 0.0010046998504549265, |
| "step": 90 |
| }, |
| { |
| "epoch": 3.1066666666666665, |
| "eval_loss": 0.001156785525381565, |
| "eval_runtime": 5.4591, |
| "eval_samples_per_second": 9.159, |
| "eval_steps_per_second": 9.159, |
| "step": 90 |
| }, |
| { |
| "epoch": 3.1066666666666665, |
| "eval_perplexity": 1.0011574548598248, |
| "step": 90 |
| }, |
| { |
| "epoch": 3.1422222222222222, |
| "grad_norm": 0.0007686283788643777, |
| "learning_rate": 7.206105506216106e-05, |
| "loss": 0.0012316190404817462, |
| "step": 91 |
| }, |
| { |
| "epoch": 3.1777777777777776, |
| "grad_norm": 0.0007238875259645283, |
| "learning_rate": 7.141808657977907e-05, |
| "loss": 0.0012630725977942348, |
| "step": 92 |
| }, |
| { |
| "epoch": 3.2133333333333334, |
| "grad_norm": 0.0005691683618351817, |
| "learning_rate": 7.077075065009433e-05, |
| "loss": 0.0010533572640269995, |
| "step": 93 |
| }, |
| { |
| "epoch": 3.2488888888888887, |
| "grad_norm": 0.0005341364885680377, |
| "learning_rate": 7.01191792739534e-05, |
| "loss": 0.0009238627390004694, |
| "step": 94 |
| }, |
| { |
| "epoch": 3.2844444444444445, |
| "grad_norm": 0.0006295841885730624, |
| "learning_rate": 6.946350531586959e-05, |
| "loss": 0.0011080841068178415, |
| "step": 95 |
| }, |
| { |
| "epoch": 3.32, |
| "grad_norm": 0.000528680335264653, |
| "learning_rate": 6.880386247692999e-05, |
| "loss": 0.001002258388325572, |
| "step": 96 |
| }, |
| { |
| "epoch": 3.3555555555555556, |
| "grad_norm": 0.0005391994491219521, |
| "learning_rate": 6.814038526753205e-05, |
| "loss": 0.0009756924118846655, |
| "step": 97 |
| }, |
| { |
| "epoch": 3.391111111111111, |
| "grad_norm": 0.0004892258439213037, |
| "learning_rate": 6.747320897995493e-05, |
| "loss": 0.0009195349994115531, |
| "step": 98 |
| }, |
| { |
| "epoch": 3.4266666666666667, |
| "grad_norm": 0.0006504295743070543, |
| "learning_rate": 6.680246966077151e-05, |
| "loss": 0.0010819350136443973, |
| "step": 99 |
| }, |
| { |
| "epoch": 3.462222222222222, |
| "grad_norm": 0.0005436805658973753, |
| "learning_rate": 6.61283040831067e-05, |
| "loss": 0.0010187672451138496, |
| "step": 100 |
| }, |
| { |
| "epoch": 3.462222222222222, |
| "eval_loss": 0.00102730724029243, |
| "eval_runtime": 5.4639, |
| "eval_samples_per_second": 9.151, |
| "eval_steps_per_second": 9.151, |
| "step": 100 |
| }, |
| { |
| "epoch": 3.462222222222222, |
| "eval_perplexity": 1.0010278351011184, |
| "step": 100 |
| }, |
| { |
| "epoch": 3.497777777777778, |
| "grad_norm": 0.0005463913548737764, |
| "learning_rate": 6.545084971874738e-05, |
| "loss": 0.0010149093577638268, |
| "step": 101 |
| }, |
| { |
| "epoch": 3.533333333333333, |
| "grad_norm": 0.0005039684474468231, |
| "learning_rate": 6.477024471011001e-05, |
| "loss": 0.0009072019602172077, |
| "step": 102 |
| }, |
| { |
| "epoch": 3.568888888888889, |
| "grad_norm": 0.0005210865056142211, |
| "learning_rate": 6.408662784207149e-05, |
| "loss": 0.0009972177213057876, |
| "step": 103 |
| }, |
| { |
| "epoch": 3.6044444444444443, |
| "grad_norm": 0.0006203350494615734, |
| "learning_rate": 6.340013851366896e-05, |
| "loss": 0.0010793538531288505, |
| "step": 104 |
| }, |
| { |
| "epoch": 3.64, |
| "grad_norm": 0.0004328833019826561, |
| "learning_rate": 6.271091670967436e-05, |
| "loss": 0.0008299812907353044, |
| "step": 105 |
| }, |
| { |
| "epoch": 3.6755555555555555, |
| "grad_norm": 0.0004699431301560253, |
| "learning_rate": 6.201910297204962e-05, |
| "loss": 0.0008918773382902145, |
| "step": 106 |
| }, |
| { |
| "epoch": 3.7111111111111112, |
| "grad_norm": 0.0005212542018853128, |
| "learning_rate": 6.132483837128823e-05, |
| "loss": 0.0008914040517993271, |
| "step": 107 |
| }, |
| { |
| "epoch": 3.7466666666666666, |
| "grad_norm": 0.000499907648190856, |
| "learning_rate": 6.062826447764883e-05, |
| "loss": 0.0009607981191948056, |
| "step": 108 |
| }, |
| { |
| "epoch": 3.7822222222222224, |
| "grad_norm": 0.00046874224790371954, |
| "learning_rate": 5.992952333228728e-05, |
| "loss": 0.000899041595403105, |
| "step": 109 |
| }, |
| { |
| "epoch": 3.8177777777777777, |
| "grad_norm": 0.00046614641905762255, |
| "learning_rate": 5.9228757418292266e-05, |
| "loss": 0.0009188801050186157, |
| "step": 110 |
| }, |
| { |
| "epoch": 3.8177777777777777, |
| "eval_loss": 0.0009438548004254699, |
| "eval_runtime": 5.47, |
| "eval_samples_per_second": 9.141, |
| "eval_steps_per_second": 9.141, |
| "step": 110 |
| }, |
| { |
| "epoch": 3.8177777777777777, |
| "eval_perplexity": 1.0009443003715415, |
| "step": 110 |
| }, |
| { |
| "epoch": 3.8177777777777777, |
| "step": 110, |
| "total_flos": 1.1595012073517875e+17, |
| "train_loss": 0.5676066864086103, |
| "train_runtime": 733.6059, |
| "train_samples_per_second": 4.907, |
| "train_steps_per_second": 0.316 |
| }, |
| { |
| "epoch": 3.8177777777777777, |
| "eval_loss": 0.0009438548004254699, |
| "eval_runtime": 5.4371, |
| "eval_samples_per_second": 9.196, |
| "eval_steps_per_second": 9.196, |
| "step": 110 |
| }, |
| { |
| "epoch": 3.8177777777777777, |
| "eval_perplexity": 1.0009443003715415, |
| "step": 110 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 232, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "save_steps": 10, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 5 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.1595012073517875e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
