| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 250, | |
| "global_step": 22156, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004513450081242101, | |
| "grad_norm": 7.729167461395264, | |
| "learning_rate": 6.768953068592058e-07, | |
| "loss": 5.9501, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.009026900162484202, | |
| "grad_norm": 13.082584381103516, | |
| "learning_rate": 1.3537906137184116e-06, | |
| "loss": 5.9447, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.013540350243726304, | |
| "grad_norm": 18.259796142578125, | |
| "learning_rate": 2.0306859205776177e-06, | |
| "loss": 5.9165, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.018053800324968405, | |
| "grad_norm": 22.386247634887695, | |
| "learning_rate": 2.7075812274368233e-06, | |
| "loss": 5.8394, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.022567250406210507, | |
| "grad_norm": 34.2584228515625, | |
| "learning_rate": 3.384476534296029e-06, | |
| "loss": 5.6527, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.022567250406210507, | |
| "eval_exact_match": 2.0056764427625353, | |
| "eval_f1": 8.992036775611602, | |
| "eval_runtime": 152.2365, | |
| "eval_samples_per_second": 70.929, | |
| "eval_steps_per_second": 17.736, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.02708070048745261, | |
| "grad_norm": 35.096343994140625, | |
| "learning_rate": 4.061371841155235e-06, | |
| "loss": 5.4208, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.03159415056869471, | |
| "grad_norm": 32.929325103759766, | |
| "learning_rate": 4.73826714801444e-06, | |
| "loss": 5.1947, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.03610760064993681, | |
| "grad_norm": 24.735565185546875, | |
| "learning_rate": 5.4151624548736465e-06, | |
| "loss": 4.9405, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.040621050731178915, | |
| "grad_norm": 22.857040405273438, | |
| "learning_rate": 6.092057761732852e-06, | |
| "loss": 4.7322, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.04513450081242101, | |
| "grad_norm": 24.466981887817383, | |
| "learning_rate": 6.768953068592058e-06, | |
| "loss": 4.4225, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04513450081242101, | |
| "eval_exact_match": 5.771050141911069, | |
| "eval_f1": 13.440407627304385, | |
| "eval_runtime": 143.7394, | |
| "eval_samples_per_second": 75.122, | |
| "eval_steps_per_second": 18.784, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04964795089366312, | |
| "grad_norm": 21.847347259521484, | |
| "learning_rate": 7.445848375451264e-06, | |
| "loss": 4.077, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.05416140097490522, | |
| "grad_norm": 27.222633361816406, | |
| "learning_rate": 8.12274368231047e-06, | |
| "loss": 3.5837, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.058674851056147316, | |
| "grad_norm": 29.45089340209961, | |
| "learning_rate": 8.799638989169675e-06, | |
| "loss": 3.0981, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.06318830113738942, | |
| "grad_norm": 29.15781593322754, | |
| "learning_rate": 9.47653429602888e-06, | |
| "loss": 2.823, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.06770175121863152, | |
| "grad_norm": 20.824087142944336, | |
| "learning_rate": 1.0153429602888087e-05, | |
| "loss": 2.4615, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.06770175121863152, | |
| "eval_exact_match": 52.82876064333018, | |
| "eval_f1": 63.363202801168775, | |
| "eval_runtime": 143.757, | |
| "eval_samples_per_second": 75.113, | |
| "eval_steps_per_second": 18.782, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.07221520129987362, | |
| "grad_norm": 33.184410095214844, | |
| "learning_rate": 1.0830324909747293e-05, | |
| "loss": 2.2565, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.07672865138111573, | |
| "grad_norm": 27.845844268798828, | |
| "learning_rate": 1.15072202166065e-05, | |
| "loss": 1.8158, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.08124210146235783, | |
| "grad_norm": 18.29555320739746, | |
| "learning_rate": 1.2184115523465704e-05, | |
| "loss": 1.7871, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.08575555154359993, | |
| "grad_norm": 23.45345687866211, | |
| "learning_rate": 1.2861010830324909e-05, | |
| "loss": 1.6184, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.09026900162484203, | |
| "grad_norm": 23.513124465942383, | |
| "learning_rate": 1.3537906137184115e-05, | |
| "loss": 1.577, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09026900162484203, | |
| "eval_exact_match": 65.42100283822138, | |
| "eval_f1": 76.31429412241118, | |
| "eval_runtime": 143.6001, | |
| "eval_samples_per_second": 75.195, | |
| "eval_steps_per_second": 18.802, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09478245170608413, | |
| "grad_norm": 13.316262245178223, | |
| "learning_rate": 1.4214801444043322e-05, | |
| "loss": 1.5352, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.09929590178732624, | |
| "grad_norm": 34.4163932800293, | |
| "learning_rate": 1.4891696750902528e-05, | |
| "loss": 1.6355, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.10380935186856834, | |
| "grad_norm": 33.572750091552734, | |
| "learning_rate": 1.5568592057761735e-05, | |
| "loss": 1.5064, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.10832280194981043, | |
| "grad_norm": 13.494843482971191, | |
| "learning_rate": 1.624548736462094e-05, | |
| "loss": 1.4047, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.11283625203105253, | |
| "grad_norm": 21.778579711914062, | |
| "learning_rate": 1.6922382671480144e-05, | |
| "loss": 1.2552, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.11283625203105253, | |
| "eval_exact_match": 69.5364238410596, | |
| "eval_f1": 80.48975233211664, | |
| "eval_runtime": 143.6325, | |
| "eval_samples_per_second": 75.178, | |
| "eval_steps_per_second": 18.798, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.11734970211229463, | |
| "grad_norm": 17.293298721313477, | |
| "learning_rate": 1.759927797833935e-05, | |
| "loss": 1.2779, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.12186315219353674, | |
| "grad_norm": 14.029269218444824, | |
| "learning_rate": 1.8276173285198557e-05, | |
| "loss": 1.2871, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.12637660227477884, | |
| "grad_norm": 12.971822738647461, | |
| "learning_rate": 1.895306859205776e-05, | |
| "loss": 1.1974, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.13089005235602094, | |
| "grad_norm": 30.38484001159668, | |
| "learning_rate": 1.9629963898916967e-05, | |
| "loss": 1.355, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.13540350243726304, | |
| "grad_norm": 29.467548370361328, | |
| "learning_rate": 2.0306859205776173e-05, | |
| "loss": 1.1713, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13540350243726304, | |
| "eval_exact_match": 72.57332071901608, | |
| "eval_f1": 82.65844387552723, | |
| "eval_runtime": 143.752, | |
| "eval_samples_per_second": 75.115, | |
| "eval_steps_per_second": 18.782, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13991695251850514, | |
| "grad_norm": 12.46554183959961, | |
| "learning_rate": 2.098375451263538e-05, | |
| "loss": 1.1677, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.14443040259974724, | |
| "grad_norm": 25.593875885009766, | |
| "learning_rate": 2.1660649819494586e-05, | |
| "loss": 1.3478, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.14894385268098934, | |
| "grad_norm": 16.136869430541992, | |
| "learning_rate": 2.2337545126353793e-05, | |
| "loss": 1.1676, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.15345730276223146, | |
| "grad_norm": 16.83846664428711, | |
| "learning_rate": 2.3014440433213e-05, | |
| "loss": 1.1741, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.15797075284347356, | |
| "grad_norm": 17.464096069335938, | |
| "learning_rate": 2.3691335740072202e-05, | |
| "loss": 1.2104, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.15797075284347356, | |
| "eval_exact_match": 74.64522232734153, | |
| "eval_f1": 84.72687223622708, | |
| "eval_runtime": 143.5946, | |
| "eval_samples_per_second": 75.198, | |
| "eval_steps_per_second": 18.803, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.16248420292471566, | |
| "grad_norm": 13.083732604980469, | |
| "learning_rate": 2.436823104693141e-05, | |
| "loss": 1.1279, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.16699765300595776, | |
| "grad_norm": 15.166335105895996, | |
| "learning_rate": 2.5045126353790615e-05, | |
| "loss": 1.1395, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.17151110308719986, | |
| "grad_norm": 8.829039573669434, | |
| "learning_rate": 2.5722021660649818e-05, | |
| "loss": 1.1374, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.17602455316844196, | |
| "grad_norm": 28.089391708374023, | |
| "learning_rate": 2.6398916967509024e-05, | |
| "loss": 1.2106, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.18053800324968405, | |
| "grad_norm": 13.704926490783691, | |
| "learning_rate": 2.707581227436823e-05, | |
| "loss": 1.2369, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.18053800324968405, | |
| "eval_exact_match": 77.96594134342479, | |
| "eval_f1": 86.52116394116426, | |
| "eval_runtime": 143.6954, | |
| "eval_samples_per_second": 75.145, | |
| "eval_steps_per_second": 18.79, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.18505145333092615, | |
| "grad_norm": 13.062108993530273, | |
| "learning_rate": 2.7752707581227437e-05, | |
| "loss": 1.0064, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.18956490341216825, | |
| "grad_norm": 25.21763801574707, | |
| "learning_rate": 2.8429602888086644e-05, | |
| "loss": 1.1127, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.19407835349341035, | |
| "grad_norm": 7.10919189453125, | |
| "learning_rate": 2.910649819494585e-05, | |
| "loss": 1.054, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.19859180357465248, | |
| "grad_norm": 14.997174263000488, | |
| "learning_rate": 2.9783393501805057e-05, | |
| "loss": 1.1187, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.20310525365589457, | |
| "grad_norm": 9.683287620544434, | |
| "learning_rate": 2.9948846539618856e-05, | |
| "loss": 0.9916, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.20310525365589457, | |
| "eval_exact_match": 77.21854304635761, | |
| "eval_f1": 85.59348158373206, | |
| "eval_runtime": 143.7109, | |
| "eval_samples_per_second": 75.137, | |
| "eval_steps_per_second": 18.788, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.20761870373713667, | |
| "grad_norm": 18.712541580200195, | |
| "learning_rate": 2.987362086258776e-05, | |
| "loss": 1.1756, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.21213215381837877, | |
| "grad_norm": 8.502235412597656, | |
| "learning_rate": 2.979839518555667e-05, | |
| "loss": 1.0872, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.21664560389962087, | |
| "grad_norm": 16.61508560180664, | |
| "learning_rate": 2.9723169508525577e-05, | |
| "loss": 1.1148, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.22115905398086297, | |
| "grad_norm": 9.676267623901367, | |
| "learning_rate": 2.9647943831494482e-05, | |
| "loss": 1.0559, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.22567250406210507, | |
| "grad_norm": 11.562779426574707, | |
| "learning_rate": 2.957271815446339e-05, | |
| "loss": 1.0628, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.22567250406210507, | |
| "eval_exact_match": 79.3755912961211, | |
| "eval_f1": 87.42406194378296, | |
| "eval_runtime": 143.6631, | |
| "eval_samples_per_second": 75.162, | |
| "eval_steps_per_second": 18.794, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.23018595414334717, | |
| "grad_norm": 20.24138832092285, | |
| "learning_rate": 2.9497492477432297e-05, | |
| "loss": 1.0549, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.23469940422458926, | |
| "grad_norm": 24.723041534423828, | |
| "learning_rate": 2.9422266800401203e-05, | |
| "loss": 1.137, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.2392128543058314, | |
| "grad_norm": 14.101241111755371, | |
| "learning_rate": 2.9347041123370113e-05, | |
| "loss": 1.0199, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.2437263043870735, | |
| "grad_norm": 7.032845497131348, | |
| "learning_rate": 2.927181544633902e-05, | |
| "loss": 1.0601, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.2482397544683156, | |
| "grad_norm": 13.543634414672852, | |
| "learning_rate": 2.9196589769307924e-05, | |
| "loss": 1.0534, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.2482397544683156, | |
| "eval_exact_match": 79.57426679280984, | |
| "eval_f1": 87.74281924363757, | |
| "eval_runtime": 143.6905, | |
| "eval_samples_per_second": 75.148, | |
| "eval_steps_per_second": 18.79, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.2527532045495577, | |
| "grad_norm": 7.890726566314697, | |
| "learning_rate": 2.9121364092276833e-05, | |
| "loss": 1.0196, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.2572666546307998, | |
| "grad_norm": 12.943625450134277, | |
| "learning_rate": 2.904613841524574e-05, | |
| "loss": 1.0178, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.2617801047120419, | |
| "grad_norm": 9.828871726989746, | |
| "learning_rate": 2.897091273821464e-05, | |
| "loss": 1.0482, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.266293554793284, | |
| "grad_norm": 28.693660736083984, | |
| "learning_rate": 2.889568706118355e-05, | |
| "loss": 0.9897, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.2708070048745261, | |
| "grad_norm": 10.408865928649902, | |
| "learning_rate": 2.8820461384152457e-05, | |
| "loss": 0.9407, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2708070048745261, | |
| "eval_exact_match": 80.37842951750237, | |
| "eval_f1": 88.60122498039404, | |
| "eval_runtime": 143.7217, | |
| "eval_samples_per_second": 75.131, | |
| "eval_steps_per_second": 18.786, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2753204549557682, | |
| "grad_norm": 11.135859489440918, | |
| "learning_rate": 2.8745235707121363e-05, | |
| "loss": 0.9192, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.2798339050370103, | |
| "grad_norm": 10.159820556640625, | |
| "learning_rate": 2.8670010030090272e-05, | |
| "loss": 0.9232, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.2843473551182524, | |
| "grad_norm": 6.99199914932251, | |
| "learning_rate": 2.8594784353059178e-05, | |
| "loss": 0.9188, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.2888608051994945, | |
| "grad_norm": 11.692395210266113, | |
| "learning_rate": 2.8519558676028083e-05, | |
| "loss": 0.979, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.2933742552807366, | |
| "grad_norm": 12.289103507995605, | |
| "learning_rate": 2.8444332998996993e-05, | |
| "loss": 0.8573, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.2933742552807366, | |
| "eval_exact_match": 81.44749290444655, | |
| "eval_f1": 88.89880962072144, | |
| "eval_runtime": 143.7737, | |
| "eval_samples_per_second": 75.104, | |
| "eval_steps_per_second": 18.78, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.29788770536197867, | |
| "grad_norm": 23.986677169799805, | |
| "learning_rate": 2.83691073219659e-05, | |
| "loss": 0.879, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.30240115544322077, | |
| "grad_norm": 10.214922904968262, | |
| "learning_rate": 2.8293881644934804e-05, | |
| "loss": 0.9526, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.3069146055244629, | |
| "grad_norm": 11.930830955505371, | |
| "learning_rate": 2.8218655967903714e-05, | |
| "loss": 1.0308, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.311428055605705, | |
| "grad_norm": 7.23566198348999, | |
| "learning_rate": 2.814343029087262e-05, | |
| "loss": 0.9849, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.3159415056869471, | |
| "grad_norm": 10.037336349487305, | |
| "learning_rate": 2.8068204613841525e-05, | |
| "loss": 0.8577, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3159415056869471, | |
| "eval_exact_match": 82.28949858088932, | |
| "eval_f1": 89.36851469763961, | |
| "eval_runtime": 143.8541, | |
| "eval_samples_per_second": 75.062, | |
| "eval_steps_per_second": 18.769, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3204549557681892, | |
| "grad_norm": 19.438182830810547, | |
| "learning_rate": 2.799297893681043e-05, | |
| "loss": 0.9586, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.3249684058494313, | |
| "grad_norm": 11.259856224060059, | |
| "learning_rate": 2.7917753259779337e-05, | |
| "loss": 0.923, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.3294818559306734, | |
| "grad_norm": 16.151851654052734, | |
| "learning_rate": 2.7842527582748243e-05, | |
| "loss": 0.9551, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.3339953060119155, | |
| "grad_norm": 12.570643424987793, | |
| "learning_rate": 2.7767301905717152e-05, | |
| "loss": 0.9377, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.3385087560931576, | |
| "grad_norm": 13.616052627563477, | |
| "learning_rate": 2.7692076228686058e-05, | |
| "loss": 0.7839, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.3385087560931576, | |
| "eval_exact_match": 81.69347209082308, | |
| "eval_f1": 89.48678563794635, | |
| "eval_runtime": 143.7031, | |
| "eval_samples_per_second": 75.141, | |
| "eval_steps_per_second": 18.789, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.3430222061743997, | |
| "grad_norm": 20.124753952026367, | |
| "learning_rate": 2.7616850551654964e-05, | |
| "loss": 0.9229, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.3475356562556418, | |
| "grad_norm": 7.816183090209961, | |
| "learning_rate": 2.7541624874623873e-05, | |
| "loss": 1.0717, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.3520491063368839, | |
| "grad_norm": 5.988482475280762, | |
| "learning_rate": 2.746639919759278e-05, | |
| "loss": 0.9611, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.356562556418126, | |
| "grad_norm": 13.177979469299316, | |
| "learning_rate": 2.7391173520561685e-05, | |
| "loss": 0.9207, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.3610760064993681, | |
| "grad_norm": 11.034092903137207, | |
| "learning_rate": 2.7315947843530594e-05, | |
| "loss": 0.9395, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3610760064993681, | |
| "eval_exact_match": 82.74361400189214, | |
| "eval_f1": 90.06638594360132, | |
| "eval_runtime": 143.7901, | |
| "eval_samples_per_second": 75.096, | |
| "eval_steps_per_second": 18.777, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3655894565806102, | |
| "grad_norm": 5.799317359924316, | |
| "learning_rate": 2.72407221664995e-05, | |
| "loss": 0.9385, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.3701029066618523, | |
| "grad_norm": 13.385774612426758, | |
| "learning_rate": 2.7165496489468405e-05, | |
| "loss": 0.9356, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.3746163567430944, | |
| "grad_norm": 22.11754608154297, | |
| "learning_rate": 2.7090270812437315e-05, | |
| "loss": 0.8532, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.3791298068243365, | |
| "grad_norm": 4.648535251617432, | |
| "learning_rate": 2.701504513540622e-05, | |
| "loss": 1.0032, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.3836432569055786, | |
| "grad_norm": 24.29154396057129, | |
| "learning_rate": 2.6939819458375123e-05, | |
| "loss": 0.8606, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.3836432569055786, | |
| "eval_exact_match": 82.58278145695364, | |
| "eval_f1": 89.65557078580815, | |
| "eval_runtime": 143.6035, | |
| "eval_samples_per_second": 75.193, | |
| "eval_steps_per_second": 18.802, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.3881567069868207, | |
| "grad_norm": 21.021242141723633, | |
| "learning_rate": 2.6864593781344032e-05, | |
| "loss": 0.9006, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.39267015706806285, | |
| "grad_norm": 26.466794967651367, | |
| "learning_rate": 2.6789368104312938e-05, | |
| "loss": 0.9426, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.39718360714930495, | |
| "grad_norm": 6.325038433074951, | |
| "learning_rate": 2.6714142427281844e-05, | |
| "loss": 0.8097, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.40169705723054705, | |
| "grad_norm": 8.038667678833008, | |
| "learning_rate": 2.6638916750250753e-05, | |
| "loss": 0.9053, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.40621050731178915, | |
| "grad_norm": 10.573040008544922, | |
| "learning_rate": 2.656369107321966e-05, | |
| "loss": 0.8459, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.40621050731178915, | |
| "eval_exact_match": 82.60170293282876, | |
| "eval_f1": 89.80136126079411, | |
| "eval_runtime": 143.5994, | |
| "eval_samples_per_second": 75.195, | |
| "eval_steps_per_second": 18.802, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.41072395739303125, | |
| "grad_norm": 7.336009979248047, | |
| "learning_rate": 2.6488465396188565e-05, | |
| "loss": 0.8691, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.41523740747427335, | |
| "grad_norm": 13.7125825881958, | |
| "learning_rate": 2.6413239719157474e-05, | |
| "loss": 0.8486, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.41975085755551544, | |
| "grad_norm": 12.19320011138916, | |
| "learning_rate": 2.633801404212638e-05, | |
| "loss": 0.9189, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.42426430763675754, | |
| "grad_norm": 11.599879264831543, | |
| "learning_rate": 2.6262788365095286e-05, | |
| "loss": 0.942, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.42877775771799964, | |
| "grad_norm": 9.139724731445312, | |
| "learning_rate": 2.6187562688064195e-05, | |
| "loss": 0.9565, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.42877775771799964, | |
| "eval_exact_match": 83.66130558183538, | |
| "eval_f1": 90.80495165338898, | |
| "eval_runtime": 143.5723, | |
| "eval_samples_per_second": 75.209, | |
| "eval_steps_per_second": 18.806, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.43329120779924174, | |
| "grad_norm": 10.977174758911133, | |
| "learning_rate": 2.61123370110331e-05, | |
| "loss": 0.8751, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.43780465788048384, | |
| "grad_norm": 13.7095947265625, | |
| "learning_rate": 2.6037111334002007e-05, | |
| "loss": 0.7547, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.44231810796172594, | |
| "grad_norm": 26.184358596801758, | |
| "learning_rate": 2.5961885656970912e-05, | |
| "loss": 0.9376, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.44683155804296804, | |
| "grad_norm": 21.620555877685547, | |
| "learning_rate": 2.5886659979939818e-05, | |
| "loss": 0.8013, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 0.45134500812421013, | |
| "grad_norm": 12.163994789123535, | |
| "learning_rate": 2.5811434302908724e-05, | |
| "loss": 0.8731, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.45134500812421013, | |
| "eval_exact_match": 82.37464522232735, | |
| "eval_f1": 89.97068346136126, | |
| "eval_runtime": 143.7072, | |
| "eval_samples_per_second": 75.139, | |
| "eval_steps_per_second": 18.788, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.45585845820545223, | |
| "grad_norm": 5.3110175132751465, | |
| "learning_rate": 2.5736208625877633e-05, | |
| "loss": 0.9062, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 0.46037190828669433, | |
| "grad_norm": 17.91867446899414, | |
| "learning_rate": 2.566098294884654e-05, | |
| "loss": 0.8749, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.46488535836793643, | |
| "grad_norm": 34.21914291381836, | |
| "learning_rate": 2.5585757271815445e-05, | |
| "loss": 0.8915, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 0.46939880844917853, | |
| "grad_norm": 8.76441478729248, | |
| "learning_rate": 2.5510531594784354e-05, | |
| "loss": 0.8023, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.4739122585304206, | |
| "grad_norm": 20.71419334411621, | |
| "learning_rate": 2.543530591775326e-05, | |
| "loss": 0.8114, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.4739122585304206, | |
| "eval_exact_match": 82.36518448438979, | |
| "eval_f1": 90.09893335252144, | |
| "eval_runtime": 143.9381, | |
| "eval_samples_per_second": 75.018, | |
| "eval_steps_per_second": 18.758, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.4784257086116628, | |
| "grad_norm": 14.544415473937988, | |
| "learning_rate": 2.536008024072217e-05, | |
| "loss": 0.8045, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.4829391586929049, | |
| "grad_norm": 40.066375732421875, | |
| "learning_rate": 2.5284854563691075e-05, | |
| "loss": 0.8651, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 0.487452608774147, | |
| "grad_norm": 16.154937744140625, | |
| "learning_rate": 2.520962888665998e-05, | |
| "loss": 0.9995, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.4919660588553891, | |
| "grad_norm": 6.674190044403076, | |
| "learning_rate": 2.513440320962889e-05, | |
| "loss": 0.8231, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 0.4964795089366312, | |
| "grad_norm": 6.037493705749512, | |
| "learning_rate": 2.5059177532597796e-05, | |
| "loss": 0.8369, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.4964795089366312, | |
| "eval_exact_match": 83.20719016083254, | |
| "eval_f1": 90.18831406264282, | |
| "eval_runtime": 143.9261, | |
| "eval_samples_per_second": 75.025, | |
| "eval_steps_per_second": 18.76, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5009929590178732, | |
| "grad_norm": 8.621197700500488, | |
| "learning_rate": 2.4983951855566702e-05, | |
| "loss": 0.8002, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 0.5055064090991154, | |
| "grad_norm": 19.25075340270996, | |
| "learning_rate": 2.4908726178535608e-05, | |
| "loss": 0.8039, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.5100198591803574, | |
| "grad_norm": 14.201600074768066, | |
| "learning_rate": 2.4833500501504514e-05, | |
| "loss": 0.7525, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 0.5145333092615996, | |
| "grad_norm": 30.636154174804688, | |
| "learning_rate": 2.475827482447342e-05, | |
| "loss": 0.7711, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.5190467593428417, | |
| "grad_norm": 8.79736042022705, | |
| "learning_rate": 2.468304914744233e-05, | |
| "loss": 0.8997, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.5190467593428417, | |
| "eval_exact_match": 83.66130558183538, | |
| "eval_f1": 90.84316221305555, | |
| "eval_runtime": 143.8756, | |
| "eval_samples_per_second": 75.051, | |
| "eval_steps_per_second": 18.766, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.5235602094240838, | |
| "grad_norm": 14.871445655822754, | |
| "learning_rate": 2.4607823470411234e-05, | |
| "loss": 0.7639, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.5280736595053259, | |
| "grad_norm": 6.112968444824219, | |
| "learning_rate": 2.453259779338014e-05, | |
| "loss": 0.8643, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 0.532587109586568, | |
| "grad_norm": 6.213535785675049, | |
| "learning_rate": 2.445737211634905e-05, | |
| "loss": 1.0582, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.5371005596678101, | |
| "grad_norm": 4.76146936416626, | |
| "learning_rate": 2.4382146439317955e-05, | |
| "loss": 0.818, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 0.5416140097490522, | |
| "grad_norm": 8.690106391906738, | |
| "learning_rate": 2.430692076228686e-05, | |
| "loss": 0.8871, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5416140097490522, | |
| "eval_exact_match": 84.03027436140019, | |
| "eval_f1": 90.88429950527104, | |
| "eval_runtime": 143.8461, | |
| "eval_samples_per_second": 75.066, | |
| "eval_steps_per_second": 18.77, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5461274598302943, | |
| "grad_norm": 18.575305938720703, | |
| "learning_rate": 2.423169508525577e-05, | |
| "loss": 0.9143, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 0.5506409099115364, | |
| "grad_norm": 2.229325294494629, | |
| "learning_rate": 2.4156469408224676e-05, | |
| "loss": 0.9387, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.5551543599927785, | |
| "grad_norm": 9.413180351257324, | |
| "learning_rate": 2.4081243731193582e-05, | |
| "loss": 0.8657, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 0.5596678100740206, | |
| "grad_norm": 5.644939422607422, | |
| "learning_rate": 2.400601805416249e-05, | |
| "loss": 0.8445, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.5641812601552627, | |
| "grad_norm": 23.247257232666016, | |
| "learning_rate": 2.3930792377131394e-05, | |
| "loss": 0.8342, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.5641812601552627, | |
| "eval_exact_match": 84.12488174077578, | |
| "eval_f1": 91.10403462345704, | |
| "eval_runtime": 144.0358, | |
| "eval_samples_per_second": 74.967, | |
| "eval_steps_per_second": 18.745, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.5686947102365048, | |
| "grad_norm": 19.933300018310547, | |
| "learning_rate": 2.38555667001003e-05, | |
| "loss": 0.8348, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.5732081603177469, | |
| "grad_norm": 7.6812872886657715, | |
| "learning_rate": 2.378034102306921e-05, | |
| "loss": 0.7582, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 0.577721610398989, | |
| "grad_norm": 13.01408863067627, | |
| "learning_rate": 2.3705115346038115e-05, | |
| "loss": 0.7245, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.5822350604802311, | |
| "grad_norm": 23.909793853759766, | |
| "learning_rate": 2.362988966900702e-05, | |
| "loss": 0.7503, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 0.5867485105614731, | |
| "grad_norm": 5.6074323654174805, | |
| "learning_rate": 2.355466399197593e-05, | |
| "loss": 0.8251, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.5867485105614731, | |
| "eval_exact_match": 84.31409649952697, | |
| "eval_f1": 91.10669916586389, | |
| "eval_runtime": 143.8718, | |
| "eval_samples_per_second": 75.053, | |
| "eval_steps_per_second": 18.767, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.5912619606427153, | |
| "grad_norm": 13.117137908935547, | |
| "learning_rate": 2.3479438314944836e-05, | |
| "loss": 0.7903, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 0.5957754107239573, | |
| "grad_norm": 4.99781608581543, | |
| "learning_rate": 2.340421263791374e-05, | |
| "loss": 0.8538, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.6002888608051995, | |
| "grad_norm": 7.639380931854248, | |
| "learning_rate": 2.332898696088265e-05, | |
| "loss": 0.8154, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 0.6048023108864415, | |
| "grad_norm": 30.98665428161621, | |
| "learning_rate": 2.3253761283851556e-05, | |
| "loss": 0.723, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.6093157609676837, | |
| "grad_norm": 30.613746643066406, | |
| "learning_rate": 2.3178535606820462e-05, | |
| "loss": 0.8682, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.6093157609676837, | |
| "eval_exact_match": 83.72753074739829, | |
| "eval_f1": 90.83640909549077, | |
| "eval_runtime": 143.634, | |
| "eval_samples_per_second": 75.177, | |
| "eval_steps_per_second": 18.798, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.6138292110489258, | |
| "grad_norm": 8.440532684326172, | |
| "learning_rate": 2.310330992978937e-05, | |
| "loss": 0.8729, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.6183426611301679, | |
| "grad_norm": 5.947940826416016, | |
| "learning_rate": 2.3028084252758277e-05, | |
| "loss": 0.747, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 0.62285611121141, | |
| "grad_norm": 16.59714698791504, | |
| "learning_rate": 2.295285857572718e-05, | |
| "loss": 0.8015, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.6273695612926521, | |
| "grad_norm": 5.211153507232666, | |
| "learning_rate": 2.287763289869609e-05, | |
| "loss": 0.8957, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 0.6318830113738942, | |
| "grad_norm": 4.547276496887207, | |
| "learning_rate": 2.2802407221664995e-05, | |
| "loss": 0.9019, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.6318830113738942, | |
| "eval_exact_match": 84.82497634815516, | |
| "eval_f1": 91.52669904904272, | |
| "eval_runtime": 145.3332, | |
| "eval_samples_per_second": 74.298, | |
| "eval_steps_per_second": 18.578, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.6363964614551363, | |
| "grad_norm": 11.642155647277832, | |
| "learning_rate": 2.27271815446339e-05, | |
| "loss": 0.7963, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 0.6409099115363784, | |
| "grad_norm": 7.39171028137207, | |
| "learning_rate": 2.265195586760281e-05, | |
| "loss": 0.7953, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.6454233616176205, | |
| "grad_norm": 13.905296325683594, | |
| "learning_rate": 2.2576730190571716e-05, | |
| "loss": 0.7865, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 0.6499368116988626, | |
| "grad_norm": 5.167139530181885, | |
| "learning_rate": 2.250150451354062e-05, | |
| "loss": 0.8087, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.6544502617801047, | |
| "grad_norm": 27.534217834472656, | |
| "learning_rate": 2.242627883650953e-05, | |
| "loss": 0.9436, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.6544502617801047, | |
| "eval_exact_match": 84.06811731315042, | |
| "eval_f1": 91.07524033930977, | |
| "eval_runtime": 143.7651, | |
| "eval_samples_per_second": 75.109, | |
| "eval_steps_per_second": 18.781, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.6589637118613468, | |
| "grad_norm": 15.742715835571289, | |
| "learning_rate": 2.2351053159478437e-05, | |
| "loss": 0.8499, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.6634771619425889, | |
| "grad_norm": 16.15327262878418, | |
| "learning_rate": 2.2275827482447342e-05, | |
| "loss": 0.846, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 0.667990612023831, | |
| "grad_norm": 17.383888244628906, | |
| "learning_rate": 2.220060180541625e-05, | |
| "loss": 0.7903, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.6725040621050731, | |
| "grad_norm": 7.484638214111328, | |
| "learning_rate": 2.2125376128385157e-05, | |
| "loss": 0.7664, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 0.6770175121863152, | |
| "grad_norm": 10.082265853881836, | |
| "learning_rate": 2.2050150451354063e-05, | |
| "loss": 0.9177, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.6770175121863152, | |
| "eval_exact_match": 84.49385052034059, | |
| "eval_f1": 91.10452090726004, | |
| "eval_runtime": 143.5023, | |
| "eval_samples_per_second": 75.246, | |
| "eval_steps_per_second": 18.815, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.6815309622675573, | |
| "grad_norm": 10.778836250305176, | |
| "learning_rate": 2.1974924774322973e-05, | |
| "loss": 0.7529, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 0.6860444123487994, | |
| "grad_norm": 12.894726753234863, | |
| "learning_rate": 2.1899699097291875e-05, | |
| "loss": 0.8783, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.6905578624300415, | |
| "grad_norm": 7.819123268127441, | |
| "learning_rate": 2.182447342026078e-05, | |
| "loss": 0.9095, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 0.6950713125112836, | |
| "grad_norm": 24.68296241760254, | |
| "learning_rate": 2.174924774322969e-05, | |
| "loss": 0.8804, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.6995847625925258, | |
| "grad_norm": 9.52649974822998, | |
| "learning_rate": 2.1674022066198596e-05, | |
| "loss": 0.8028, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.6995847625925258, | |
| "eval_exact_match": 85.59129612109744, | |
| "eval_f1": 91.93623152881347, | |
| "eval_runtime": 143.4073, | |
| "eval_samples_per_second": 75.296, | |
| "eval_steps_per_second": 18.827, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.7040982126737678, | |
| "grad_norm": 10.898487091064453, | |
| "learning_rate": 2.1598796389167502e-05, | |
| "loss": 0.8282, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.70861166275501, | |
| "grad_norm": 6.693902969360352, | |
| "learning_rate": 2.152357071213641e-05, | |
| "loss": 0.775, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 0.713125112836252, | |
| "grad_norm": 10.00558090209961, | |
| "learning_rate": 2.1448345035105317e-05, | |
| "loss": 0.6894, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.7176385629174942, | |
| "grad_norm": 2.776298761367798, | |
| "learning_rate": 2.1373119358074223e-05, | |
| "loss": 0.8409, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 0.7221520129987362, | |
| "grad_norm": 13.581101417541504, | |
| "learning_rate": 2.1297893681043132e-05, | |
| "loss": 0.8222, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7221520129987362, | |
| "eval_exact_match": 84.57899716177862, | |
| "eval_f1": 91.38107826122027, | |
| "eval_runtime": 143.4879, | |
| "eval_samples_per_second": 75.254, | |
| "eval_steps_per_second": 18.817, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7266654630799784, | |
| "grad_norm": 6.5702223777771, | |
| "learning_rate": 2.1222668004012038e-05, | |
| "loss": 0.6735, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 0.7311789131612204, | |
| "grad_norm": 18.275623321533203, | |
| "learning_rate": 2.1147442326980944e-05, | |
| "loss": 0.8389, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.7356923632424626, | |
| "grad_norm": 15.205418586730957, | |
| "learning_rate": 2.1072216649949853e-05, | |
| "loss": 0.7803, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 0.7402058133237046, | |
| "grad_norm": 8.31666088104248, | |
| "learning_rate": 2.099699097291876e-05, | |
| "loss": 0.7081, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.7447192634049468, | |
| "grad_norm": 9.174483299255371, | |
| "learning_rate": 2.092176529588766e-05, | |
| "loss": 0.826, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.7447192634049468, | |
| "eval_exact_match": 84.76821192052981, | |
| "eval_f1": 91.56620229706857, | |
| "eval_runtime": 143.4859, | |
| "eval_samples_per_second": 75.255, | |
| "eval_steps_per_second": 18.817, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.7492327134861888, | |
| "grad_norm": 5.849365234375, | |
| "learning_rate": 2.084653961885657e-05, | |
| "loss": 0.7826, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.753746163567431, | |
| "grad_norm": 8.80666446685791, | |
| "learning_rate": 2.0771313941825476e-05, | |
| "loss": 0.8931, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 0.758259613648673, | |
| "grad_norm": 7.301697731018066, | |
| "learning_rate": 2.0696088264794382e-05, | |
| "loss": 0.6788, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.7627730637299152, | |
| "grad_norm": 9.519810676574707, | |
| "learning_rate": 2.062086258776329e-05, | |
| "loss": 0.7928, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 0.7672865138111572, | |
| "grad_norm": 8.138936996459961, | |
| "learning_rate": 2.0545636910732197e-05, | |
| "loss": 0.8625, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.7672865138111572, | |
| "eval_exact_match": 85.19394512771996, | |
| "eval_f1": 91.95119129750337, | |
| "eval_runtime": 143.5272, | |
| "eval_samples_per_second": 75.233, | |
| "eval_steps_per_second": 18.812, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.7717999638923994, | |
| "grad_norm": 3.691103935241699, | |
| "learning_rate": 2.0470411233701103e-05, | |
| "loss": 0.7947, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 0.7763134139736414, | |
| "grad_norm": 14.496338844299316, | |
| "learning_rate": 2.0395185556670012e-05, | |
| "loss": 0.8135, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.7808268640548836, | |
| "grad_norm": 6.248403072357178, | |
| "learning_rate": 2.0319959879638918e-05, | |
| "loss": 0.8594, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 0.7853403141361257, | |
| "grad_norm": 16.819801330566406, | |
| "learning_rate": 2.0244734202607824e-05, | |
| "loss": 0.8171, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.7898537642173677, | |
| "grad_norm": 7.9061079025268555, | |
| "learning_rate": 2.0169508525576733e-05, | |
| "loss": 0.6517, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.7898537642173677, | |
| "eval_exact_match": 85.6480605487228, | |
| "eval_f1": 91.95890910573651, | |
| "eval_runtime": 143.4989, | |
| "eval_samples_per_second": 75.248, | |
| "eval_steps_per_second": 18.815, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.7943672142986099, | |
| "grad_norm": 12.129390716552734, | |
| "learning_rate": 2.009428284854564e-05, | |
| "loss": 0.7358, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.798880664379852, | |
| "grad_norm": 7.113585472106934, | |
| "learning_rate": 2.0019057171514545e-05, | |
| "loss": 0.8517, | |
| "step": 8850 | |
| }, | |
| { | |
| "epoch": 0.8033941144610941, | |
| "grad_norm": 10.407898902893066, | |
| "learning_rate": 1.994383149448345e-05, | |
| "loss": 0.733, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.8079075645423361, | |
| "grad_norm": 15.745281219482422, | |
| "learning_rate": 1.9868605817452356e-05, | |
| "loss": 0.693, | |
| "step": 8950 | |
| }, | |
| { | |
| "epoch": 0.8124210146235783, | |
| "grad_norm": 6.876597881317139, | |
| "learning_rate": 1.9793380140421262e-05, | |
| "loss": 0.817, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8124210146235783, | |
| "eval_exact_match": 84.87228003784296, | |
| "eval_f1": 91.5110749584356, | |
| "eval_runtime": 143.5544, | |
| "eval_samples_per_second": 75.219, | |
| "eval_steps_per_second": 18.808, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8169344647048203, | |
| "grad_norm": 7.4037065505981445, | |
| "learning_rate": 1.971815446339017e-05, | |
| "loss": 0.8677, | |
| "step": 9050 | |
| }, | |
| { | |
| "epoch": 0.8214479147860625, | |
| "grad_norm": 4.559969902038574, | |
| "learning_rate": 1.9642928786359077e-05, | |
| "loss": 0.7798, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.8259613648673045, | |
| "grad_norm": 7.184974670410156, | |
| "learning_rate": 1.9567703109327983e-05, | |
| "loss": 0.7705, | |
| "step": 9150 | |
| }, | |
| { | |
| "epoch": 0.8304748149485467, | |
| "grad_norm": 8.206283569335938, | |
| "learning_rate": 1.9492477432296892e-05, | |
| "loss": 0.8398, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.8349882650297887, | |
| "grad_norm": 7.29602575302124, | |
| "learning_rate": 1.9417251755265798e-05, | |
| "loss": 0.7574, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 0.8349882650297887, | |
| "eval_exact_match": 85.06149479659413, | |
| "eval_f1": 91.53106503540634, | |
| "eval_runtime": 143.5779, | |
| "eval_samples_per_second": 75.207, | |
| "eval_steps_per_second": 18.805, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 0.8395017151110309, | |
| "grad_norm": 10.167183876037598, | |
| "learning_rate": 1.9342026078234704e-05, | |
| "loss": 0.7103, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 0.8440151651922729, | |
| "grad_norm": 6.271793365478516, | |
| "learning_rate": 1.9266800401203613e-05, | |
| "loss": 0.8364, | |
| "step": 9350 | |
| }, | |
| { | |
| "epoch": 0.8485286152735151, | |
| "grad_norm": 11.07026481628418, | |
| "learning_rate": 1.919157472417252e-05, | |
| "loss": 0.726, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.8530420653547571, | |
| "grad_norm": 5.571475028991699, | |
| "learning_rate": 1.9116349047141425e-05, | |
| "loss": 0.7205, | |
| "step": 9450 | |
| }, | |
| { | |
| "epoch": 0.8575555154359993, | |
| "grad_norm": 3.4866223335266113, | |
| "learning_rate": 1.9041123370110334e-05, | |
| "loss": 0.7832, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.8575555154359993, | |
| "eval_exact_match": 85.44938505203406, | |
| "eval_f1": 91.84507576310226, | |
| "eval_runtime": 143.5379, | |
| "eval_samples_per_second": 75.228, | |
| "eval_steps_per_second": 18.81, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.8620689655172413, | |
| "grad_norm": 2.2408883571624756, | |
| "learning_rate": 1.896589769307924e-05, | |
| "loss": 0.7533, | |
| "step": 9550 | |
| }, | |
| { | |
| "epoch": 0.8665824155984835, | |
| "grad_norm": 13.415377616882324, | |
| "learning_rate": 1.8890672016048142e-05, | |
| "loss": 0.7669, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.8710958656797256, | |
| "grad_norm": 4.730581760406494, | |
| "learning_rate": 1.881544633901705e-05, | |
| "loss": 0.7468, | |
| "step": 9650 | |
| }, | |
| { | |
| "epoch": 0.8756093157609677, | |
| "grad_norm": 6.725691318511963, | |
| "learning_rate": 1.8740220661985957e-05, | |
| "loss": 0.7426, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 0.8801227658422098, | |
| "grad_norm": 8.169360160827637, | |
| "learning_rate": 1.8664994984954863e-05, | |
| "loss": 0.8436, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 0.8801227658422098, | |
| "eval_exact_match": 84.88174077578051, | |
| "eval_f1": 91.83275837323971, | |
| "eval_runtime": 143.4694, | |
| "eval_samples_per_second": 75.263, | |
| "eval_steps_per_second": 18.819, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 0.8846362159234519, | |
| "grad_norm": 1.717469334602356, | |
| "learning_rate": 1.8589769307923772e-05, | |
| "loss": 0.7889, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.889149666004694, | |
| "grad_norm": 20.31835174560547, | |
| "learning_rate": 1.8514543630892678e-05, | |
| "loss": 0.7648, | |
| "step": 9850 | |
| }, | |
| { | |
| "epoch": 0.8936631160859361, | |
| "grad_norm": 15.77481746673584, | |
| "learning_rate": 1.8439317953861584e-05, | |
| "loss": 0.7259, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 0.8981765661671782, | |
| "grad_norm": 3.87709641456604, | |
| "learning_rate": 1.8364092276830493e-05, | |
| "loss": 0.7866, | |
| "step": 9950 | |
| }, | |
| { | |
| "epoch": 0.9026900162484203, | |
| "grad_norm": 8.835536003112793, | |
| "learning_rate": 1.82888665997994e-05, | |
| "loss": 0.6775, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9026900162484203, | |
| "eval_exact_match": 85.99810785241249, | |
| "eval_f1": 91.99639894905705, | |
| "eval_runtime": 143.5348, | |
| "eval_samples_per_second": 75.229, | |
| "eval_steps_per_second": 18.811, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9072034663296624, | |
| "grad_norm": 12.733137130737305, | |
| "learning_rate": 1.8213640922768305e-05, | |
| "loss": 0.7703, | |
| "step": 10050 | |
| }, | |
| { | |
| "epoch": 0.9117169164109045, | |
| "grad_norm": 12.40443229675293, | |
| "learning_rate": 1.8138415245737214e-05, | |
| "loss": 0.7619, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 0.9162303664921466, | |
| "grad_norm": 6.346498012542725, | |
| "learning_rate": 1.806318956870612e-05, | |
| "loss": 0.7605, | |
| "step": 10150 | |
| }, | |
| { | |
| "epoch": 0.9207438165733887, | |
| "grad_norm": 5.372687816619873, | |
| "learning_rate": 1.7987963891675026e-05, | |
| "loss": 0.6591, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.9252572666546308, | |
| "grad_norm": 4.377304553985596, | |
| "learning_rate": 1.7912738214643932e-05, | |
| "loss": 0.8404, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 0.9252572666546308, | |
| "eval_exact_match": 85.89403973509934, | |
| "eval_f1": 92.11360231029698, | |
| "eval_runtime": 144.2084, | |
| "eval_samples_per_second": 74.878, | |
| "eval_steps_per_second": 18.723, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 0.9297707167358729, | |
| "grad_norm": 11.62856388092041, | |
| "learning_rate": 1.7837512537612838e-05, | |
| "loss": 0.7712, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 0.934284166817115, | |
| "grad_norm": 5.723257541656494, | |
| "learning_rate": 1.7762286860581743e-05, | |
| "loss": 0.7171, | |
| "step": 10350 | |
| }, | |
| { | |
| "epoch": 0.9387976168983571, | |
| "grad_norm": 6.060873031616211, | |
| "learning_rate": 1.7687061183550653e-05, | |
| "loss": 0.7324, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.9433110669795992, | |
| "grad_norm": 4.51533842086792, | |
| "learning_rate": 1.761183550651956e-05, | |
| "loss": 0.7633, | |
| "step": 10450 | |
| }, | |
| { | |
| "epoch": 0.9478245170608413, | |
| "grad_norm": 11.809548377990723, | |
| "learning_rate": 1.7536609829488464e-05, | |
| "loss": 0.8111, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.9478245170608413, | |
| "eval_exact_match": 85.09933774834437, | |
| "eval_f1": 91.93045017438146, | |
| "eval_runtime": 143.4235, | |
| "eval_samples_per_second": 75.288, | |
| "eval_steps_per_second": 18.825, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.9523379671420834, | |
| "grad_norm": 15.76356029510498, | |
| "learning_rate": 1.7461384152457374e-05, | |
| "loss": 0.7013, | |
| "step": 10550 | |
| }, | |
| { | |
| "epoch": 0.9568514172233256, | |
| "grad_norm": 23.272687911987305, | |
| "learning_rate": 1.738615847542628e-05, | |
| "loss": 0.8229, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.9613648673045676, | |
| "grad_norm": 16.758358001708984, | |
| "learning_rate": 1.7310932798395185e-05, | |
| "loss": 0.7479, | |
| "step": 10650 | |
| }, | |
| { | |
| "epoch": 0.9658783173858098, | |
| "grad_norm": 14.670035362243652, | |
| "learning_rate": 1.7235707121364094e-05, | |
| "loss": 0.8, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 0.9703917674670518, | |
| "grad_norm": 5.1286821365356445, | |
| "learning_rate": 1.7160481444333e-05, | |
| "loss": 0.8522, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 0.9703917674670518, | |
| "eval_exact_match": 85.07095553453169, | |
| "eval_f1": 92.09316544794538, | |
| "eval_runtime": 143.6698, | |
| "eval_samples_per_second": 75.158, | |
| "eval_steps_per_second": 18.793, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 0.974905217548294, | |
| "grad_norm": 8.131464004516602, | |
| "learning_rate": 1.7085255767301906e-05, | |
| "loss": 0.7949, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.979418667629536, | |
| "grad_norm": 22.16661834716797, | |
| "learning_rate": 1.7010030090270815e-05, | |
| "loss": 0.7486, | |
| "step": 10850 | |
| }, | |
| { | |
| "epoch": 0.9839321177107782, | |
| "grad_norm": 5.554388046264648, | |
| "learning_rate": 1.693480441323972e-05, | |
| "loss": 0.7604, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 0.9884455677920202, | |
| "grad_norm": 8.525761604309082, | |
| "learning_rate": 1.6859578736208624e-05, | |
| "loss": 0.766, | |
| "step": 10950 | |
| }, | |
| { | |
| "epoch": 0.9929590178732624, | |
| "grad_norm": 10.504690170288086, | |
| "learning_rate": 1.6784353059177533e-05, | |
| "loss": 0.7166, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.9929590178732624, | |
| "eval_exact_match": 85.04257332071901, | |
| "eval_f1": 92.02241474371678, | |
| "eval_runtime": 143.4783, | |
| "eval_samples_per_second": 75.259, | |
| "eval_steps_per_second": 18.818, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.9974724679545044, | |
| "grad_norm": 7.378440856933594, | |
| "learning_rate": 1.670912738214644e-05, | |
| "loss": 0.7187, | |
| "step": 11050 | |
| }, | |
| { | |
| "epoch": 1.0019859180357464, | |
| "grad_norm": 3.172842502593994, | |
| "learning_rate": 1.6633901705115345e-05, | |
| "loss": 0.7532, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 1.0064993681169887, | |
| "grad_norm": 2.3299856185913086, | |
| "learning_rate": 1.6558676028084254e-05, | |
| "loss": 0.497, | |
| "step": 11150 | |
| }, | |
| { | |
| "epoch": 1.0110128181982307, | |
| "grad_norm": 8.6509428024292, | |
| "learning_rate": 1.648345035105316e-05, | |
| "loss": 0.4497, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 1.0155262682794728, | |
| "grad_norm": 9.68758773803711, | |
| "learning_rate": 1.6408224674022065e-05, | |
| "loss": 0.6154, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 1.0155262682794728, | |
| "eval_exact_match": 86.08325449385052, | |
| "eval_f1": 92.38528194318762, | |
| "eval_runtime": 143.3576, | |
| "eval_samples_per_second": 75.322, | |
| "eval_steps_per_second": 18.834, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 1.0200397183607148, | |
| "grad_norm": 3.34212064743042, | |
| "learning_rate": 1.6332998996990975e-05, | |
| "loss": 0.5372, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 1.024553168441957, | |
| "grad_norm": 5.384337425231934, | |
| "learning_rate": 1.625777331995988e-05, | |
| "loss": 0.5464, | |
| "step": 11350 | |
| }, | |
| { | |
| "epoch": 1.0290666185231991, | |
| "grad_norm": 19.279573440551758, | |
| "learning_rate": 1.6182547642928786e-05, | |
| "loss": 0.5558, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 1.0335800686044412, | |
| "grad_norm": 5.5248308181762695, | |
| "learning_rate": 1.6107321965897696e-05, | |
| "loss": 0.4981, | |
| "step": 11450 | |
| }, | |
| { | |
| "epoch": 1.0380935186856832, | |
| "grad_norm": 5.657703399658203, | |
| "learning_rate": 1.60320962888666e-05, | |
| "loss": 0.6565, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.0380935186856832, | |
| "eval_exact_match": 85.58183538315988, | |
| "eval_f1": 92.11318103014378, | |
| "eval_runtime": 143.5023, | |
| "eval_samples_per_second": 75.246, | |
| "eval_steps_per_second": 18.815, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.0426069687669255, | |
| "grad_norm": 6.387887954711914, | |
| "learning_rate": 1.5956870611835507e-05, | |
| "loss": 0.588, | |
| "step": 11550 | |
| }, | |
| { | |
| "epoch": 1.0471204188481675, | |
| "grad_norm": 1.7305879592895508, | |
| "learning_rate": 1.5881644934804413e-05, | |
| "loss": 0.5747, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 1.0516338689294096, | |
| "grad_norm": 14.716680526733398, | |
| "learning_rate": 1.580641925777332e-05, | |
| "loss": 0.5632, | |
| "step": 11650 | |
| }, | |
| { | |
| "epoch": 1.0561473190106518, | |
| "grad_norm": 9.127685546875, | |
| "learning_rate": 1.5731193580742225e-05, | |
| "loss": 0.4897, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 1.0606607690918939, | |
| "grad_norm": 8.541461944580078, | |
| "learning_rate": 1.5655967903711134e-05, | |
| "loss": 0.544, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 1.0606607690918939, | |
| "eval_exact_match": 86.16840113528855, | |
| "eval_f1": 92.442978713336, | |
| "eval_runtime": 143.2851, | |
| "eval_samples_per_second": 75.36, | |
| "eval_steps_per_second": 18.844, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 1.065174219173136, | |
| "grad_norm": 6.538851737976074, | |
| "learning_rate": 1.558074222668004e-05, | |
| "loss": 0.5202, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 1.069687669254378, | |
| "grad_norm": 7.314679145812988, | |
| "learning_rate": 1.5505516549648946e-05, | |
| "loss": 0.54, | |
| "step": 11850 | |
| }, | |
| { | |
| "epoch": 1.0742011193356202, | |
| "grad_norm": 2.3385446071624756, | |
| "learning_rate": 1.5430290872617855e-05, | |
| "loss": 0.5192, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 1.0787145694168623, | |
| "grad_norm": 26.518877029418945, | |
| "learning_rate": 1.535506519558676e-05, | |
| "loss": 0.5435, | |
| "step": 11950 | |
| }, | |
| { | |
| "epoch": 1.0832280194981043, | |
| "grad_norm": 39.6591682434082, | |
| "learning_rate": 1.5279839518555667e-05, | |
| "loss": 0.5149, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0832280194981043, | |
| "eval_exact_match": 85.80889309366131, | |
| "eval_f1": 92.20769990556119, | |
| "eval_runtime": 143.6087, | |
| "eval_samples_per_second": 75.19, | |
| "eval_steps_per_second": 18.801, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0877414695793464, | |
| "grad_norm": 11.38036823272705, | |
| "learning_rate": 1.5204613841524576e-05, | |
| "loss": 0.526, | |
| "step": 12050 | |
| }, | |
| { | |
| "epoch": 1.0922549196605886, | |
| "grad_norm": 21.02750587463379, | |
| "learning_rate": 1.512938816449348e-05, | |
| "loss": 0.471, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 1.0967683697418307, | |
| "grad_norm": 23.10146713256836, | |
| "learning_rate": 1.5054162487462386e-05, | |
| "loss": 0.6643, | |
| "step": 12150 | |
| }, | |
| { | |
| "epoch": 1.1012818198230727, | |
| "grad_norm": 29.241615295410156, | |
| "learning_rate": 1.4978936810431293e-05, | |
| "loss": 0.5231, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 1.1057952699043148, | |
| "grad_norm": 3.3990285396575928, | |
| "learning_rate": 1.49037111334002e-05, | |
| "loss": 0.4051, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 1.1057952699043148, | |
| "eval_exact_match": 85.56291390728477, | |
| "eval_f1": 92.0566633980034, | |
| "eval_runtime": 143.183, | |
| "eval_samples_per_second": 75.414, | |
| "eval_steps_per_second": 18.857, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 1.110308719985557, | |
| "grad_norm": 3.075737237930298, | |
| "learning_rate": 1.4828485456369108e-05, | |
| "loss": 0.5637, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 1.114822170066799, | |
| "grad_norm": 2.9517650604248047, | |
| "learning_rate": 1.4753259779338014e-05, | |
| "loss": 0.566, | |
| "step": 12350 | |
| }, | |
| { | |
| "epoch": 1.119335620148041, | |
| "grad_norm": 23.89853858947754, | |
| "learning_rate": 1.4678034102306922e-05, | |
| "loss": 0.4439, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 1.1238490702292832, | |
| "grad_norm": 8.486159324645996, | |
| "learning_rate": 1.4602808425275828e-05, | |
| "loss": 0.5538, | |
| "step": 12450 | |
| }, | |
| { | |
| "epoch": 1.1283625203105254, | |
| "grad_norm": 3.5648648738861084, | |
| "learning_rate": 1.4527582748244733e-05, | |
| "loss": 0.5173, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.1283625203105254, | |
| "eval_exact_match": 85.4872280037843, | |
| "eval_f1": 92.15147631309604, | |
| "eval_runtime": 143.2081, | |
| "eval_samples_per_second": 75.401, | |
| "eval_steps_per_second": 18.854, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.1328759703917675, | |
| "grad_norm": 10.259268760681152, | |
| "learning_rate": 1.4452357071213641e-05, | |
| "loss": 0.5925, | |
| "step": 12550 | |
| }, | |
| { | |
| "epoch": 1.1373894204730095, | |
| "grad_norm": 6.570536136627197, | |
| "learning_rate": 1.4377131394182548e-05, | |
| "loss": 0.4594, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 1.1419028705542518, | |
| "grad_norm": 6.687112808227539, | |
| "learning_rate": 1.4301905717151454e-05, | |
| "loss": 0.4994, | |
| "step": 12650 | |
| }, | |
| { | |
| "epoch": 1.1464163206354938, | |
| "grad_norm": 14.550410270690918, | |
| "learning_rate": 1.4226680040120362e-05, | |
| "loss": 0.5775, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 1.1509297707167359, | |
| "grad_norm": 12.998605728149414, | |
| "learning_rate": 1.4151454363089268e-05, | |
| "loss": 0.5285, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 1.1509297707167359, | |
| "eval_exact_match": 85.93188268684958, | |
| "eval_f1": 92.11716297833141, | |
| "eval_runtime": 143.1084, | |
| "eval_samples_per_second": 75.453, | |
| "eval_steps_per_second": 18.867, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 1.155443220797978, | |
| "grad_norm": 26.985210418701172, | |
| "learning_rate": 1.4076228686058175e-05, | |
| "loss": 0.5092, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 1.1599566708792202, | |
| "grad_norm": 15.445883750915527, | |
| "learning_rate": 1.4001003009027081e-05, | |
| "loss": 0.5178, | |
| "step": 12850 | |
| }, | |
| { | |
| "epoch": 1.1644701209604622, | |
| "grad_norm": 8.596466064453125, | |
| "learning_rate": 1.3925777331995989e-05, | |
| "loss": 0.5742, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 1.1689835710417043, | |
| "grad_norm": 3.9060676097869873, | |
| "learning_rate": 1.3850551654964896e-05, | |
| "loss": 0.5112, | |
| "step": 12950 | |
| }, | |
| { | |
| "epoch": 1.1734970211229463, | |
| "grad_norm": 2.3090436458587646, | |
| "learning_rate": 1.3775325977933802e-05, | |
| "loss": 0.4802, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.1734970211229463, | |
| "eval_exact_match": 86.3670766319773, | |
| "eval_f1": 92.39148643540621, | |
| "eval_runtime": 143.1331, | |
| "eval_samples_per_second": 75.44, | |
| "eval_steps_per_second": 18.864, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.1780104712041886, | |
| "grad_norm": 4.289682865142822, | |
| "learning_rate": 1.370010030090271e-05, | |
| "loss": 0.4555, | |
| "step": 13050 | |
| }, | |
| { | |
| "epoch": 1.1825239212854306, | |
| "grad_norm": 23.45159149169922, | |
| "learning_rate": 1.3624874623871615e-05, | |
| "loss": 0.6034, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 1.1870373713666726, | |
| "grad_norm": 14.170953750610352, | |
| "learning_rate": 1.3549648946840521e-05, | |
| "loss": 0.4946, | |
| "step": 13150 | |
| }, | |
| { | |
| "epoch": 1.191550821447915, | |
| "grad_norm": 7.408278942108154, | |
| "learning_rate": 1.3474423269809429e-05, | |
| "loss": 0.5625, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 1.196064271529157, | |
| "grad_norm": 4.187251567840576, | |
| "learning_rate": 1.3399197592778336e-05, | |
| "loss": 0.5344, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 1.196064271529157, | |
| "eval_exact_match": 85.66698202459791, | |
| "eval_f1": 92.29328625942796, | |
| "eval_runtime": 143.2334, | |
| "eval_samples_per_second": 75.387, | |
| "eval_steps_per_second": 18.85, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 1.200577721610399, | |
| "grad_norm": 9.739165306091309, | |
| "learning_rate": 1.3323971915747242e-05, | |
| "loss": 0.5319, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 1.205091171691641, | |
| "grad_norm": 3.0962629318237305, | |
| "learning_rate": 1.324874623871615e-05, | |
| "loss": 0.5455, | |
| "step": 13350 | |
| }, | |
| { | |
| "epoch": 1.209604621772883, | |
| "grad_norm": 10.260982513427734, | |
| "learning_rate": 1.3173520561685057e-05, | |
| "loss": 0.5922, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 1.2141180718541253, | |
| "grad_norm": 17.95406150817871, | |
| "learning_rate": 1.3098294884653961e-05, | |
| "loss": 0.6416, | |
| "step": 13450 | |
| }, | |
| { | |
| "epoch": 1.2186315219353674, | |
| "grad_norm": 9.253098487854004, | |
| "learning_rate": 1.3023069207622869e-05, | |
| "loss": 0.4543, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.2186315219353674, | |
| "eval_exact_match": 86.20624408703878, | |
| "eval_f1": 92.32664235875168, | |
| "eval_runtime": 143.2002, | |
| "eval_samples_per_second": 75.405, | |
| "eval_steps_per_second": 18.855, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.2231449720166094, | |
| "grad_norm": 9.202949523925781, | |
| "learning_rate": 1.2947843530591776e-05, | |
| "loss": 0.6569, | |
| "step": 13550 | |
| }, | |
| { | |
| "epoch": 1.2276584220978517, | |
| "grad_norm": 16.244760513305664, | |
| "learning_rate": 1.2872617853560682e-05, | |
| "loss": 0.5605, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 1.2321718721790937, | |
| "grad_norm": 2.6242430210113525, | |
| "learning_rate": 1.279739217652959e-05, | |
| "loss": 0.546, | |
| "step": 13650 | |
| }, | |
| { | |
| "epoch": 1.2366853222603358, | |
| "grad_norm": 19.960708618164062, | |
| "learning_rate": 1.2722166499498497e-05, | |
| "loss": 0.5916, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 1.2411987723415778, | |
| "grad_norm": 14.39201545715332, | |
| "learning_rate": 1.2646940822467401e-05, | |
| "loss": 0.4647, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 1.2411987723415778, | |
| "eval_exact_match": 86.27246925260171, | |
| "eval_f1": 92.45848778749898, | |
| "eval_runtime": 143.112, | |
| "eval_samples_per_second": 75.451, | |
| "eval_steps_per_second": 18.866, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 1.24571222242282, | |
| "grad_norm": 8.58752155303955, | |
| "learning_rate": 1.2571715145436309e-05, | |
| "loss": 0.5224, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 1.2502256725040621, | |
| "grad_norm": 5.419035911560059, | |
| "learning_rate": 1.2496489468405216e-05, | |
| "loss": 0.6743, | |
| "step": 13850 | |
| }, | |
| { | |
| "epoch": 1.2547391225853042, | |
| "grad_norm": 7.52559232711792, | |
| "learning_rate": 1.2421263791374122e-05, | |
| "loss": 0.5955, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 1.2592525726665462, | |
| "grad_norm": 15.449511528015137, | |
| "learning_rate": 1.234603811434303e-05, | |
| "loss": 0.5824, | |
| "step": 13950 | |
| }, | |
| { | |
| "epoch": 1.2637660227477885, | |
| "grad_norm": 11.266414642333984, | |
| "learning_rate": 1.2270812437311937e-05, | |
| "loss": 0.5101, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.2637660227477885, | |
| "eval_exact_match": 86.12109744560075, | |
| "eval_f1": 92.2146350604068, | |
| "eval_runtime": 145.9619, | |
| "eval_samples_per_second": 73.978, | |
| "eval_steps_per_second": 18.498, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.2682794728290305, | |
| "grad_norm": 7.292428970336914, | |
| "learning_rate": 1.2195586760280843e-05, | |
| "loss": 0.4962, | |
| "step": 14050 | |
| }, | |
| { | |
| "epoch": 1.2727929229102726, | |
| "grad_norm": 1.1534169912338257, | |
| "learning_rate": 1.2120361083249749e-05, | |
| "loss": 0.5691, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 1.2773063729915148, | |
| "grad_norm": 3.5983633995056152, | |
| "learning_rate": 1.2045135406218656e-05, | |
| "loss": 0.5114, | |
| "step": 14150 | |
| }, | |
| { | |
| "epoch": 1.2818198230727569, | |
| "grad_norm": 5.006545543670654, | |
| "learning_rate": 1.1969909729187562e-05, | |
| "loss": 0.5259, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 1.286333273153999, | |
| "grad_norm": 24.31420135498047, | |
| "learning_rate": 1.189468405215647e-05, | |
| "loss": 0.4771, | |
| "step": 14250 | |
| }, | |
| { | |
| "epoch": 1.286333273153999, | |
| "eval_exact_match": 86.59413434247871, | |
| "eval_f1": 92.49677313517446, | |
| "eval_runtime": 145.8932, | |
| "eval_samples_per_second": 74.013, | |
| "eval_steps_per_second": 18.507, | |
| "step": 14250 | |
| }, | |
| { | |
| "epoch": 1.290846723235241, | |
| "grad_norm": 8.447436332702637, | |
| "learning_rate": 1.1819458375125377e-05, | |
| "loss": 0.5835, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 1.295360173316483, | |
| "grad_norm": 23.178955078125, | |
| "learning_rate": 1.1744232698094283e-05, | |
| "loss": 0.5554, | |
| "step": 14350 | |
| }, | |
| { | |
| "epoch": 1.2998736233977253, | |
| "grad_norm": 16.500057220458984, | |
| "learning_rate": 1.166900702106319e-05, | |
| "loss": 0.4928, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 1.3043870734789673, | |
| "grad_norm": 8.389457702636719, | |
| "learning_rate": 1.1593781344032097e-05, | |
| "loss": 0.6872, | |
| "step": 14450 | |
| }, | |
| { | |
| "epoch": 1.3089005235602094, | |
| "grad_norm": 5.315954685211182, | |
| "learning_rate": 1.1518555667001002e-05, | |
| "loss": 0.5394, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.3089005235602094, | |
| "eval_exact_match": 86.45222327341533, | |
| "eval_f1": 92.58788732745475, | |
| "eval_runtime": 145.7277, | |
| "eval_samples_per_second": 74.097, | |
| "eval_steps_per_second": 18.528, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.3134139736414516, | |
| "grad_norm": 2.0151515007019043, | |
| "learning_rate": 1.144332998996991e-05, | |
| "loss": 0.4745, | |
| "step": 14550 | |
| }, | |
| { | |
| "epoch": 1.3179274237226937, | |
| "grad_norm": 6.995370864868164, | |
| "learning_rate": 1.1368104312938817e-05, | |
| "loss": 0.536, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 1.3224408738039357, | |
| "grad_norm": 4.453261852264404, | |
| "learning_rate": 1.1292878635907723e-05, | |
| "loss": 0.4909, | |
| "step": 14650 | |
| }, | |
| { | |
| "epoch": 1.3269543238851778, | |
| "grad_norm": 3.472259998321533, | |
| "learning_rate": 1.121765295887663e-05, | |
| "loss": 0.586, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 1.3314677739664198, | |
| "grad_norm": 15.908103942871094, | |
| "learning_rate": 1.1142427281845537e-05, | |
| "loss": 0.528, | |
| "step": 14750 | |
| }, | |
| { | |
| "epoch": 1.3314677739664198, | |
| "eval_exact_match": 85.86565752128666, | |
| "eval_f1": 92.39225966105154, | |
| "eval_runtime": 280.4149, | |
| "eval_samples_per_second": 38.507, | |
| "eval_steps_per_second": 9.629, | |
| "step": 14750 | |
| }, | |
| { | |
| "epoch": 1.335981224047662, | |
| "grad_norm": 12.543098449707031, | |
| "learning_rate": 1.1067201604814443e-05, | |
| "loss": 0.4912, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 1.340494674128904, | |
| "grad_norm": 24.144222259521484, | |
| "learning_rate": 1.099197592778335e-05, | |
| "loss": 0.5626, | |
| "step": 14850 | |
| }, | |
| { | |
| "epoch": 1.3450081242101461, | |
| "grad_norm": 25.347875595092773, | |
| "learning_rate": 1.0916750250752258e-05, | |
| "loss": 0.4955, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 1.3495215742913884, | |
| "grad_norm": 5.940708637237549, | |
| "learning_rate": 1.0841524573721163e-05, | |
| "loss": 0.4859, | |
| "step": 14950 | |
| }, | |
| { | |
| "epoch": 1.3540350243726305, | |
| "grad_norm": 30.9013671875, | |
| "learning_rate": 1.0766298896690071e-05, | |
| "loss": 0.4788, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.3540350243726305, | |
| "eval_exact_match": 86.6414380321665, | |
| "eval_f1": 92.62032707644155, | |
| "eval_runtime": 145.6846, | |
| "eval_samples_per_second": 74.119, | |
| "eval_steps_per_second": 18.533, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.3585484744538725, | |
| "grad_norm": 12.216713905334473, | |
| "learning_rate": 1.0691073219658978e-05, | |
| "loss": 0.4977, | |
| "step": 15050 | |
| }, | |
| { | |
| "epoch": 1.3630619245351148, | |
| "grad_norm": 40.2611083984375, | |
| "learning_rate": 1.0615847542627883e-05, | |
| "loss": 0.4972, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 1.3675753746163568, | |
| "grad_norm": 10.3711519241333, | |
| "learning_rate": 1.054062186559679e-05, | |
| "loss": 0.4955, | |
| "step": 15150 | |
| }, | |
| { | |
| "epoch": 1.3720888246975989, | |
| "grad_norm": 2.400322914123535, | |
| "learning_rate": 1.0465396188565698e-05, | |
| "loss": 0.4868, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 1.376602274778841, | |
| "grad_norm": 3.9988925457000732, | |
| "learning_rate": 1.0390170511534603e-05, | |
| "loss": 0.5888, | |
| "step": 15250 | |
| }, | |
| { | |
| "epoch": 1.376602274778841, | |
| "eval_exact_match": 85.96026490066225, | |
| "eval_f1": 92.33197764854948, | |
| "eval_runtime": 145.4185, | |
| "eval_samples_per_second": 74.255, | |
| "eval_steps_per_second": 18.567, | |
| "step": 15250 | |
| }, | |
| { | |
| "epoch": 1.381115724860083, | |
| "grad_norm": 1.6575514078140259, | |
| "learning_rate": 1.0314944834503511e-05, | |
| "loss": 0.5143, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 1.3856291749413252, | |
| "grad_norm": 5.943323612213135, | |
| "learning_rate": 1.0239719157472419e-05, | |
| "loss": 0.4725, | |
| "step": 15350 | |
| }, | |
| { | |
| "epoch": 1.3901426250225672, | |
| "grad_norm": 21.014570236206055, | |
| "learning_rate": 1.0164493480441324e-05, | |
| "loss": 0.5131, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 1.3946560751038093, | |
| "grad_norm": 4.148115634918213, | |
| "learning_rate": 1.008926780341023e-05, | |
| "loss": 0.4323, | |
| "step": 15450 | |
| }, | |
| { | |
| "epoch": 1.3991695251850516, | |
| "grad_norm": 8.95993423461914, | |
| "learning_rate": 1.0014042126379138e-05, | |
| "loss": 0.6072, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.3991695251850516, | |
| "eval_exact_match": 85.76158940397352, | |
| "eval_f1": 92.24988076673156, | |
| "eval_runtime": 145.3259, | |
| "eval_samples_per_second": 74.302, | |
| "eval_steps_per_second": 18.579, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.4036829752662936, | |
| "grad_norm": 5.6876959800720215, | |
| "learning_rate": 9.938816449348044e-06, | |
| "loss": 0.5692, | |
| "step": 15550 | |
| }, | |
| { | |
| "epoch": 1.4081964253475356, | |
| "grad_norm": 6.91029167175293, | |
| "learning_rate": 9.863590772316951e-06, | |
| "loss": 0.5801, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 1.4127098754287777, | |
| "grad_norm": 8.116116523742676, | |
| "learning_rate": 9.788365095285859e-06, | |
| "loss": 0.4394, | |
| "step": 15650 | |
| }, | |
| { | |
| "epoch": 1.4172233255100197, | |
| "grad_norm": 7.001738548278809, | |
| "learning_rate": 9.713139418254764e-06, | |
| "loss": 0.5607, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 1.421736775591262, | |
| "grad_norm": 21.804443359375, | |
| "learning_rate": 9.637913741223672e-06, | |
| "loss": 0.5191, | |
| "step": 15750 | |
| }, | |
| { | |
| "epoch": 1.421736775591262, | |
| "eval_exact_match": 85.97918637653737, | |
| "eval_f1": 92.37275066667881, | |
| "eval_runtime": 145.3503, | |
| "eval_samples_per_second": 74.289, | |
| "eval_steps_per_second": 18.576, | |
| "step": 15750 | |
| }, | |
| { | |
| "epoch": 1.426250225672504, | |
| "grad_norm": 11.133319854736328, | |
| "learning_rate": 9.562688064192578e-06, | |
| "loss": 0.5256, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 1.430763675753746, | |
| "grad_norm": 5.4904632568359375, | |
| "learning_rate": 9.487462387161484e-06, | |
| "loss": 0.5177, | |
| "step": 15850 | |
| }, | |
| { | |
| "epoch": 1.4352771258349883, | |
| "grad_norm": 9.791414260864258, | |
| "learning_rate": 9.412236710130391e-06, | |
| "loss": 0.5814, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 1.4397905759162304, | |
| "grad_norm": 3.3400447368621826, | |
| "learning_rate": 9.337011033099299e-06, | |
| "loss": 0.5374, | |
| "step": 15950 | |
| }, | |
| { | |
| "epoch": 1.4443040259974724, | |
| "grad_norm": 23.98038673400879, | |
| "learning_rate": 9.261785356068205e-06, | |
| "loss": 0.4973, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.4443040259974724, | |
| "eval_exact_match": 84.85335856196784, | |
| "eval_f1": 92.03377983249271, | |
| "eval_runtime": 145.2676, | |
| "eval_samples_per_second": 74.332, | |
| "eval_steps_per_second": 18.586, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.4488174760787147, | |
| "grad_norm": 9.59720516204834, | |
| "learning_rate": 9.186559679037112e-06, | |
| "loss": 0.534, | |
| "step": 16050 | |
| }, | |
| { | |
| "epoch": 1.4533309261599567, | |
| "grad_norm": 10.079476356506348, | |
| "learning_rate": 9.111334002006018e-06, | |
| "loss": 0.511, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 1.4578443762411988, | |
| "grad_norm": 3.377192497253418, | |
| "learning_rate": 9.036108324974924e-06, | |
| "loss": 0.5187, | |
| "step": 16150 | |
| }, | |
| { | |
| "epoch": 1.4623578263224408, | |
| "grad_norm": 10.79287052154541, | |
| "learning_rate": 8.960882647943831e-06, | |
| "loss": 0.5499, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 1.4668712764036829, | |
| "grad_norm": 15.751055717468262, | |
| "learning_rate": 8.885656970912739e-06, | |
| "loss": 0.536, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 1.4668712764036829, | |
| "eval_exact_match": 86.40491958372753, | |
| "eval_f1": 92.48354485469106, | |
| "eval_runtime": 145.2653, | |
| "eval_samples_per_second": 74.333, | |
| "eval_steps_per_second": 18.587, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 1.4713847264849251, | |
| "grad_norm": 3.9903676509857178, | |
| "learning_rate": 8.810431293881645e-06, | |
| "loss": 0.5385, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 1.4758981765661672, | |
| "grad_norm": 6.5839080810546875, | |
| "learning_rate": 8.735205616850552e-06, | |
| "loss": 0.5092, | |
| "step": 16350 | |
| }, | |
| { | |
| "epoch": 1.4804116266474092, | |
| "grad_norm": 13.69189453125, | |
| "learning_rate": 8.65997993981946e-06, | |
| "loss": 0.5999, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 1.4849250767286515, | |
| "grad_norm": 11.840332984924316, | |
| "learning_rate": 8.584754262788364e-06, | |
| "loss": 0.57, | |
| "step": 16450 | |
| }, | |
| { | |
| "epoch": 1.4894385268098935, | |
| "grad_norm": 11.86502742767334, | |
| "learning_rate": 8.509528585757271e-06, | |
| "loss": 0.4635, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.4894385268098935, | |
| "eval_exact_match": 85.49668874172185, | |
| "eval_f1": 92.28753097702375, | |
| "eval_runtime": 145.2047, | |
| "eval_samples_per_second": 74.364, | |
| "eval_steps_per_second": 18.594, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.4939519768911356, | |
| "grad_norm": 17.741037368774414, | |
| "learning_rate": 8.434302908726179e-06, | |
| "loss": 0.53, | |
| "step": 16550 | |
| }, | |
| { | |
| "epoch": 1.4984654269723776, | |
| "grad_norm": 7.774323463439941, | |
| "learning_rate": 8.359077231695085e-06, | |
| "loss": 0.575, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 1.5029788770536197, | |
| "grad_norm": 4.973544597625732, | |
| "learning_rate": 8.283851554663992e-06, | |
| "loss": 0.6468, | |
| "step": 16650 | |
| }, | |
| { | |
| "epoch": 1.507492327134862, | |
| "grad_norm": 5.228555202484131, | |
| "learning_rate": 8.2086258776329e-06, | |
| "loss": 0.4901, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 1.512005777216104, | |
| "grad_norm": 3.4082319736480713, | |
| "learning_rate": 8.133400200601806e-06, | |
| "loss": 0.5492, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 1.512005777216104, | |
| "eval_exact_match": 86.10217596972564, | |
| "eval_f1": 92.39499925755595, | |
| "eval_runtime": 145.1347, | |
| "eval_samples_per_second": 74.4, | |
| "eval_steps_per_second": 18.603, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 1.516519227297346, | |
| "grad_norm": 20.74472427368164, | |
| "learning_rate": 8.058174523570712e-06, | |
| "loss": 0.4685, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 1.5210326773785883, | |
| "grad_norm": 4.805381774902344, | |
| "learning_rate": 7.982948846539619e-06, | |
| "loss": 0.474, | |
| "step": 16850 | |
| }, | |
| { | |
| "epoch": 1.5255461274598303, | |
| "grad_norm": 1.6686218976974487, | |
| "learning_rate": 7.907723169508525e-06, | |
| "loss": 0.5617, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 1.5300595775410724, | |
| "grad_norm": 14.378780364990234, | |
| "learning_rate": 7.832497492477432e-06, | |
| "loss": 0.4829, | |
| "step": 16950 | |
| }, | |
| { | |
| "epoch": 1.5345730276223146, | |
| "grad_norm": 9.25706672668457, | |
| "learning_rate": 7.75727181544634e-06, | |
| "loss": 0.5525, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.5345730276223146, | |
| "eval_exact_match": 85.89403973509934, | |
| "eval_f1": 92.2175102581889, | |
| "eval_runtime": 145.4454, | |
| "eval_samples_per_second": 74.241, | |
| "eval_steps_per_second": 18.564, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.5390864777035564, | |
| "grad_norm": 10.210553169250488, | |
| "learning_rate": 7.682046138415246e-06, | |
| "loss": 0.5284, | |
| "step": 17050 | |
| }, | |
| { | |
| "epoch": 1.5435999277847987, | |
| "grad_norm": 18.55254364013672, | |
| "learning_rate": 7.6068204613841525e-06, | |
| "loss": 0.4863, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 1.5481133778660408, | |
| "grad_norm": 9.640850067138672, | |
| "learning_rate": 7.53159478435306e-06, | |
| "loss": 0.6163, | |
| "step": 17150 | |
| }, | |
| { | |
| "epoch": 1.5526268279472828, | |
| "grad_norm": 7.999804496765137, | |
| "learning_rate": 7.456369107321966e-06, | |
| "loss": 0.518, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 1.557140278028525, | |
| "grad_norm": 9.162345886230469, | |
| "learning_rate": 7.3811434302908725e-06, | |
| "loss": 0.5001, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 1.557140278028525, | |
| "eval_exact_match": 85.9035004730369, | |
| "eval_f1": 92.33731500742522, | |
| "eval_runtime": 145.3659, | |
| "eval_samples_per_second": 74.282, | |
| "eval_steps_per_second": 18.574, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 1.561653728109767, | |
| "grad_norm": 5.462348461151123, | |
| "learning_rate": 7.30591775325978e-06, | |
| "loss": 0.4179, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 1.5661671781910091, | |
| "grad_norm": 17.67523765563965, | |
| "learning_rate": 7.230692076228686e-06, | |
| "loss": 0.4662, | |
| "step": 17350 | |
| }, | |
| { | |
| "epoch": 1.5706806282722514, | |
| "grad_norm": 4.397737503051758, | |
| "learning_rate": 7.155466399197593e-06, | |
| "loss": 0.4614, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 1.5751940783534935, | |
| "grad_norm": 7.665886402130127, | |
| "learning_rate": 7.0802407221665e-06, | |
| "loss": 0.5263, | |
| "step": 17450 | |
| }, | |
| { | |
| "epoch": 1.5797075284347355, | |
| "grad_norm": 10.627632141113281, | |
| "learning_rate": 7.005015045135407e-06, | |
| "loss": 0.4021, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.5797075284347355, | |
| "eval_exact_match": 85.98864711447493, | |
| "eval_f1": 92.40008411758966, | |
| "eval_runtime": 145.3304, | |
| "eval_samples_per_second": 74.3, | |
| "eval_steps_per_second": 18.578, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.5842209785159778, | |
| "grad_norm": 7.057243347167969, | |
| "learning_rate": 6.929789368104313e-06, | |
| "loss": 0.5326, | |
| "step": 17550 | |
| }, | |
| { | |
| "epoch": 1.5887344285972196, | |
| "grad_norm": 8.216778755187988, | |
| "learning_rate": 6.85456369107322e-06, | |
| "loss": 0.6029, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 1.5932478786784618, | |
| "grad_norm": 3.943422794342041, | |
| "learning_rate": 6.779338014042127e-06, | |
| "loss": 0.5518, | |
| "step": 17650 | |
| }, | |
| { | |
| "epoch": 1.597761328759704, | |
| "grad_norm": 12.350107192993164, | |
| "learning_rate": 6.704112337011033e-06, | |
| "loss": 0.5368, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 1.602274778840946, | |
| "grad_norm": 6.516546249389648, | |
| "learning_rate": 6.62888665997994e-06, | |
| "loss": 0.504, | |
| "step": 17750 | |
| }, | |
| { | |
| "epoch": 1.602274778840946, | |
| "eval_exact_match": 86.12109744560075, | |
| "eval_f1": 92.37865780721518, | |
| "eval_runtime": 145.3649, | |
| "eval_samples_per_second": 74.282, | |
| "eval_steps_per_second": 18.574, | |
| "step": 17750 | |
| }, | |
| { | |
| "epoch": 1.6067882289221882, | |
| "grad_norm": 3.5462801456451416, | |
| "learning_rate": 6.553660982948847e-06, | |
| "loss": 0.4613, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 1.6113016790034302, | |
| "grad_norm": 9.32242488861084, | |
| "learning_rate": 6.478435305917753e-06, | |
| "loss": 0.5149, | |
| "step": 17850 | |
| }, | |
| { | |
| "epoch": 1.6158151290846723, | |
| "grad_norm": 4.5879597663879395, | |
| "learning_rate": 6.40320962888666e-06, | |
| "loss": 0.4189, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 1.6203285791659146, | |
| "grad_norm": 10.474478721618652, | |
| "learning_rate": 6.327983951855567e-06, | |
| "loss": 0.5182, | |
| "step": 17950 | |
| }, | |
| { | |
| "epoch": 1.6248420292471564, | |
| "grad_norm": 4.693137168884277, | |
| "learning_rate": 6.252758274824474e-06, | |
| "loss": 0.6212, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.6248420292471564, | |
| "eval_exact_match": 85.93188268684958, | |
| "eval_f1": 92.33148266916612, | |
| "eval_runtime": 145.2483, | |
| "eval_samples_per_second": 74.342, | |
| "eval_steps_per_second": 18.589, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.6293554793283986, | |
| "grad_norm": 20.36956214904785, | |
| "learning_rate": 6.17753259779338e-06, | |
| "loss": 0.6194, | |
| "step": 18050 | |
| }, | |
| { | |
| "epoch": 1.6338689294096407, | |
| "grad_norm": 8.450358390808105, | |
| "learning_rate": 6.102306920762287e-06, | |
| "loss": 0.518, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 1.6383823794908827, | |
| "grad_norm": 4.681309223175049, | |
| "learning_rate": 6.027081243731194e-06, | |
| "loss": 0.4853, | |
| "step": 18150 | |
| }, | |
| { | |
| "epoch": 1.642895829572125, | |
| "grad_norm": 6.480415344238281, | |
| "learning_rate": 5.9518555667001e-06, | |
| "loss": 0.5101, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 1.647409279653367, | |
| "grad_norm": 11.254326820373535, | |
| "learning_rate": 5.876629889669007e-06, | |
| "loss": 0.5095, | |
| "step": 18250 | |
| }, | |
| { | |
| "epoch": 1.647409279653367, | |
| "eval_exact_match": 86.03595080416272, | |
| "eval_f1": 92.38608711058883, | |
| "eval_runtime": 145.2161, | |
| "eval_samples_per_second": 74.358, | |
| "eval_steps_per_second": 18.593, | |
| "step": 18250 | |
| }, | |
| { | |
| "epoch": 1.651922729734609, | |
| "grad_norm": 7.776529788970947, | |
| "learning_rate": 5.801404212637914e-06, | |
| "loss": 0.5322, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 1.6564361798158513, | |
| "grad_norm": 17.068607330322266, | |
| "learning_rate": 5.72617853560682e-06, | |
| "loss": 0.4755, | |
| "step": 18350 | |
| }, | |
| { | |
| "epoch": 1.6609496298970934, | |
| "grad_norm": 20.472034454345703, | |
| "learning_rate": 5.650952858575727e-06, | |
| "loss": 0.6017, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 1.6654630799783354, | |
| "grad_norm": 4.904719352722168, | |
| "learning_rate": 5.575727181544634e-06, | |
| "loss": 0.4662, | |
| "step": 18450 | |
| }, | |
| { | |
| "epoch": 1.6699765300595777, | |
| "grad_norm": 7.219258785247803, | |
| "learning_rate": 5.500501504513541e-06, | |
| "loss": 0.4781, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.6699765300595777, | |
| "eval_exact_match": 85.55345316934721, | |
| "eval_f1": 92.32101574735367, | |
| "eval_runtime": 145.1791, | |
| "eval_samples_per_second": 74.377, | |
| "eval_steps_per_second": 18.598, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.6744899801408195, | |
| "grad_norm": 8.811306953430176, | |
| "learning_rate": 5.425275827482447e-06, | |
| "loss": 0.5161, | |
| "step": 18550 | |
| }, | |
| { | |
| "epoch": 1.6790034302220618, | |
| "grad_norm": 4.903675079345703, | |
| "learning_rate": 5.350050150451354e-06, | |
| "loss": 0.5215, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 1.6835168803033038, | |
| "grad_norm": 2.371656894683838, | |
| "learning_rate": 5.274824473420261e-06, | |
| "loss": 0.486, | |
| "step": 18650 | |
| }, | |
| { | |
| "epoch": 1.6880303303845459, | |
| "grad_norm": 8.991338729858398, | |
| "learning_rate": 5.199598796389167e-06, | |
| "loss": 0.5043, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 1.6925437804657881, | |
| "grad_norm": 5.460509777069092, | |
| "learning_rate": 5.124373119358074e-06, | |
| "loss": 0.5344, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 1.6925437804657881, | |
| "eval_exact_match": 86.02649006622516, | |
| "eval_f1": 92.44635703301584, | |
| "eval_runtime": 145.2855, | |
| "eval_samples_per_second": 74.323, | |
| "eval_steps_per_second": 18.584, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 1.6970572305470302, | |
| "grad_norm": 6.287936210632324, | |
| "learning_rate": 5.049147442326981e-06, | |
| "loss": 0.4446, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 1.7015706806282722, | |
| "grad_norm": 2.3766534328460693, | |
| "learning_rate": 4.973921765295887e-06, | |
| "loss": 0.4618, | |
| "step": 18850 | |
| }, | |
| { | |
| "epoch": 1.7060841307095145, | |
| "grad_norm": 6.606088161468506, | |
| "learning_rate": 4.898696088264794e-06, | |
| "loss": 0.498, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 1.7105975807907563, | |
| "grad_norm": 7.917613506317139, | |
| "learning_rate": 4.8234704112337015e-06, | |
| "loss": 0.473, | |
| "step": 18950 | |
| }, | |
| { | |
| "epoch": 1.7151110308719986, | |
| "grad_norm": 13.437002182006836, | |
| "learning_rate": 4.748244734202608e-06, | |
| "loss": 0.5217, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.7151110308719986, | |
| "eval_exact_match": 86.2251655629139, | |
| "eval_f1": 92.51045927542914, | |
| "eval_runtime": 145.2525, | |
| "eval_samples_per_second": 74.34, | |
| "eval_steps_per_second": 18.588, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.7196244809532406, | |
| "grad_norm": 21.756275177001953, | |
| "learning_rate": 4.673019057171515e-06, | |
| "loss": 0.5129, | |
| "step": 19050 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 4.2581377029418945, | |
| "learning_rate": 4.5977933801404215e-06, | |
| "loss": 0.5202, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 1.728651381115725, | |
| "grad_norm": 9.396230697631836, | |
| "learning_rate": 4.522567703109328e-06, | |
| "loss": 0.5118, | |
| "step": 19150 | |
| }, | |
| { | |
| "epoch": 1.733164831196967, | |
| "grad_norm": 9.545235633850098, | |
| "learning_rate": 4.447342026078235e-06, | |
| "loss": 0.4611, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 1.737678281278209, | |
| "grad_norm": 4.104794502258301, | |
| "learning_rate": 4.3721163490471416e-06, | |
| "loss": 0.5737, | |
| "step": 19250 | |
| }, | |
| { | |
| "epoch": 1.737678281278209, | |
| "eval_exact_match": 86.23462630085146, | |
| "eval_f1": 92.57135940815057, | |
| "eval_runtime": 145.5739, | |
| "eval_samples_per_second": 74.175, | |
| "eval_steps_per_second": 18.547, | |
| "step": 19250 | |
| }, | |
| { | |
| "epoch": 1.7421917313594513, | |
| "grad_norm": 8.624117851257324, | |
| "learning_rate": 4.296890672016048e-06, | |
| "loss": 0.5349, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 1.7467051814406933, | |
| "grad_norm": 4.802499771118164, | |
| "learning_rate": 4.221664994984955e-06, | |
| "loss": 0.5332, | |
| "step": 19350 | |
| }, | |
| { | |
| "epoch": 1.7512186315219354, | |
| "grad_norm": 4.347715854644775, | |
| "learning_rate": 4.146439317953862e-06, | |
| "loss": 0.474, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 1.7557320816031776, | |
| "grad_norm": 21.51348114013672, | |
| "learning_rate": 4.071213640922768e-06, | |
| "loss": 0.5182, | |
| "step": 19450 | |
| }, | |
| { | |
| "epoch": 1.7602455316844194, | |
| "grad_norm": 3.002976655960083, | |
| "learning_rate": 3.995987963891676e-06, | |
| "loss": 0.4243, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.7602455316844194, | |
| "eval_exact_match": 86.44276253547777, | |
| "eval_f1": 92.67282094003843, | |
| "eval_runtime": 145.268, | |
| "eval_samples_per_second": 74.332, | |
| "eval_steps_per_second": 18.586, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.7647589817656617, | |
| "grad_norm": 6.619145393371582, | |
| "learning_rate": 3.920762286860582e-06, | |
| "loss": 0.5754, | |
| "step": 19550 | |
| }, | |
| { | |
| "epoch": 1.7692724318469037, | |
| "grad_norm": 8.654962539672852, | |
| "learning_rate": 3.845536609829488e-06, | |
| "loss": 0.4927, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 1.7737858819281458, | |
| "grad_norm": 2.102865695953369, | |
| "learning_rate": 3.7703109327983955e-06, | |
| "loss": 0.417, | |
| "step": 19650 | |
| }, | |
| { | |
| "epoch": 1.778299332009388, | |
| "grad_norm": 9.824490547180176, | |
| "learning_rate": 3.695085255767302e-06, | |
| "loss": 0.5777, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 1.78281278209063, | |
| "grad_norm": 22.286598205566406, | |
| "learning_rate": 3.619859578736209e-06, | |
| "loss": 0.4338, | |
| "step": 19750 | |
| }, | |
| { | |
| "epoch": 1.78281278209063, | |
| "eval_exact_match": 86.39545884578997, | |
| "eval_f1": 92.63887659164942, | |
| "eval_runtime": 145.3006, | |
| "eval_samples_per_second": 74.315, | |
| "eval_steps_per_second": 18.582, | |
| "step": 19750 | |
| }, | |
| { | |
| "epoch": 1.7873262321718721, | |
| "grad_norm": 0.6572410464286804, | |
| "learning_rate": 3.5446339017051155e-06, | |
| "loss": 0.4549, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 1.7918396822531144, | |
| "grad_norm": 15.171038627624512, | |
| "learning_rate": 3.469408224674022e-06, | |
| "loss": 0.452, | |
| "step": 19850 | |
| }, | |
| { | |
| "epoch": 1.7963531323343562, | |
| "grad_norm": 13.550349235534668, | |
| "learning_rate": 3.394182547642929e-06, | |
| "loss": 0.4901, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 1.8008665824155985, | |
| "grad_norm": 9.970057487487793, | |
| "learning_rate": 3.3189568706118356e-06, | |
| "loss": 0.4424, | |
| "step": 19950 | |
| }, | |
| { | |
| "epoch": 1.8053800324968405, | |
| "grad_norm": 3.251477003097534, | |
| "learning_rate": 3.2437311935807422e-06, | |
| "loss": 0.5214, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.8053800324968405, | |
| "eval_exact_match": 86.10217596972564, | |
| "eval_f1": 92.55287274359681, | |
| "eval_runtime": 145.2611, | |
| "eval_samples_per_second": 74.335, | |
| "eval_steps_per_second": 18.587, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.8098934825780826, | |
| "grad_norm": 3.656310558319092, | |
| "learning_rate": 3.1685055165496493e-06, | |
| "loss": 0.4794, | |
| "step": 20050 | |
| }, | |
| { | |
| "epoch": 1.8144069326593248, | |
| "grad_norm": 6.139503479003906, | |
| "learning_rate": 3.0932798395185556e-06, | |
| "loss": 0.4485, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 1.8189203827405669, | |
| "grad_norm": 6.566440582275391, | |
| "learning_rate": 3.0180541624874623e-06, | |
| "loss": 0.456, | |
| "step": 20150 | |
| }, | |
| { | |
| "epoch": 1.823433832821809, | |
| "grad_norm": 6.406381130218506, | |
| "learning_rate": 2.9428284854563694e-06, | |
| "loss": 0.4988, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 1.8279472829030512, | |
| "grad_norm": 4.750673770904541, | |
| "learning_rate": 2.8676028084252757e-06, | |
| "loss": 0.5102, | |
| "step": 20250 | |
| }, | |
| { | |
| "epoch": 1.8279472829030512, | |
| "eval_exact_match": 86.0643330179754, | |
| "eval_f1": 92.48790625031062, | |
| "eval_runtime": 145.6532, | |
| "eval_samples_per_second": 74.135, | |
| "eval_steps_per_second": 18.537, | |
| "step": 20250 | |
| }, | |
| { | |
| "epoch": 1.8324607329842932, | |
| "grad_norm": 2.9004476070404053, | |
| "learning_rate": 2.7923771313941828e-06, | |
| "loss": 0.4172, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 1.8369741830655353, | |
| "grad_norm": 5.453982353210449, | |
| "learning_rate": 2.7171514543630894e-06, | |
| "loss": 0.4328, | |
| "step": 20350 | |
| }, | |
| { | |
| "epoch": 1.8414876331467775, | |
| "grad_norm": 6.562243461608887, | |
| "learning_rate": 2.6419257773319957e-06, | |
| "loss": 0.5946, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 1.8460010832280194, | |
| "grad_norm": 2.186967134475708, | |
| "learning_rate": 2.566700100300903e-06, | |
| "loss": 0.5174, | |
| "step": 20450 | |
| }, | |
| { | |
| "epoch": 1.8505145333092616, | |
| "grad_norm": 18.740962982177734, | |
| "learning_rate": 2.4914744232698095e-06, | |
| "loss": 0.5614, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.8505145333092616, | |
| "eval_exact_match": 86.24408703878902, | |
| "eval_f1": 92.6215979315234, | |
| "eval_runtime": 145.5466, | |
| "eval_samples_per_second": 74.189, | |
| "eval_steps_per_second": 18.551, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.8550279833905037, | |
| "grad_norm": 8.845901489257812, | |
| "learning_rate": 2.416248746238716e-06, | |
| "loss": 0.5174, | |
| "step": 20550 | |
| }, | |
| { | |
| "epoch": 1.8595414334717457, | |
| "grad_norm": 2.0935425758361816, | |
| "learning_rate": 2.341023069207623e-06, | |
| "loss": 0.4191, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 1.864054883552988, | |
| "grad_norm": 5.225878715515137, | |
| "learning_rate": 2.2657973921765295e-06, | |
| "loss": 0.5168, | |
| "step": 20650 | |
| }, | |
| { | |
| "epoch": 1.86856833363423, | |
| "grad_norm": 3.618779182434082, | |
| "learning_rate": 2.1905717151454362e-06, | |
| "loss": 0.5116, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 1.873081783715472, | |
| "grad_norm": 14.580885887145996, | |
| "learning_rate": 2.115346038114343e-06, | |
| "loss": 0.5247, | |
| "step": 20750 | |
| }, | |
| { | |
| "epoch": 1.873081783715472, | |
| "eval_exact_match": 86.39545884578997, | |
| "eval_f1": 92.68082620123108, | |
| "eval_runtime": 144.0826, | |
| "eval_samples_per_second": 74.943, | |
| "eval_steps_per_second": 18.739, | |
| "step": 20750 | |
| }, | |
| { | |
| "epoch": 1.8775952337967143, | |
| "grad_norm": 6.323169708251953, | |
| "learning_rate": 2.04012036108325e-06, | |
| "loss": 0.5007, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 1.8821086838779562, | |
| "grad_norm": 7.055742263793945, | |
| "learning_rate": 1.9648946840521567e-06, | |
| "loss": 0.5387, | |
| "step": 20850 | |
| }, | |
| { | |
| "epoch": 1.8866221339591984, | |
| "grad_norm": 6.097321033477783, | |
| "learning_rate": 1.8896690070210632e-06, | |
| "loss": 0.4795, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 1.8911355840404405, | |
| "grad_norm": 20.577049255371094, | |
| "learning_rate": 1.81444332998997e-06, | |
| "loss": 0.5636, | |
| "step": 20950 | |
| }, | |
| { | |
| "epoch": 1.8956490341216825, | |
| "grad_norm": 11.891510009765625, | |
| "learning_rate": 1.7392176529588768e-06, | |
| "loss": 0.5768, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.8956490341216825, | |
| "eval_exact_match": 86.28192999053927, | |
| "eval_f1": 92.59481897569101, | |
| "eval_runtime": 143.5237, | |
| "eval_samples_per_second": 75.235, | |
| "eval_steps_per_second": 18.812, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.9001624842029248, | |
| "grad_norm": 28.808475494384766, | |
| "learning_rate": 1.6639919759277832e-06, | |
| "loss": 0.5267, | |
| "step": 21050 | |
| }, | |
| { | |
| "epoch": 1.9046759342841668, | |
| "grad_norm": 11.045042991638184, | |
| "learning_rate": 1.5887662988966901e-06, | |
| "loss": 0.5508, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 1.9091893843654089, | |
| "grad_norm": 4.862325191497803, | |
| "learning_rate": 1.5135406218655968e-06, | |
| "loss": 0.4697, | |
| "step": 21150 | |
| }, | |
| { | |
| "epoch": 1.9137028344466511, | |
| "grad_norm": 13.387544631958008, | |
| "learning_rate": 1.4383149448345037e-06, | |
| "loss": 0.4416, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 1.9182162845278932, | |
| "grad_norm": 5.2306342124938965, | |
| "learning_rate": 1.3630892678034104e-06, | |
| "loss": 0.4124, | |
| "step": 21250 | |
| }, | |
| { | |
| "epoch": 1.9182162845278932, | |
| "eval_exact_match": 86.26300851466415, | |
| "eval_f1": 92.60255176769716, | |
| "eval_runtime": 143.5894, | |
| "eval_samples_per_second": 75.201, | |
| "eval_steps_per_second": 18.804, | |
| "step": 21250 | |
| }, | |
| { | |
| "epoch": 1.9227297346091352, | |
| "grad_norm": 29.35004234313965, | |
| "learning_rate": 1.2878635907723169e-06, | |
| "loss": 0.5344, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 1.9272431846903775, | |
| "grad_norm": 8.634255409240723, | |
| "learning_rate": 1.2126379137412237e-06, | |
| "loss": 0.4815, | |
| "step": 21350 | |
| }, | |
| { | |
| "epoch": 1.9317566347716193, | |
| "grad_norm": 8.262895584106445, | |
| "learning_rate": 1.1374122367101304e-06, | |
| "loss": 0.4939, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 1.9362700848528616, | |
| "grad_norm": 12.539655685424805, | |
| "learning_rate": 1.0621865596790371e-06, | |
| "loss": 0.4957, | |
| "step": 21450 | |
| }, | |
| { | |
| "epoch": 1.9407835349341036, | |
| "grad_norm": 9.728516578674316, | |
| "learning_rate": 9.86960882647944e-07, | |
| "loss": 0.4587, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.9407835349341036, | |
| "eval_exact_match": 86.14001892147587, | |
| "eval_f1": 92.57583651293868, | |
| "eval_runtime": 143.5952, | |
| "eval_samples_per_second": 75.197, | |
| "eval_steps_per_second": 18.803, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.9452969850153456, | |
| "grad_norm": 9.588170051574707, | |
| "learning_rate": 9.117352056168506e-07, | |
| "loss": 0.5395, | |
| "step": 21550 | |
| }, | |
| { | |
| "epoch": 1.949810435096588, | |
| "grad_norm": 14.394529342651367, | |
| "learning_rate": 8.365095285857573e-07, | |
| "loss": 0.5462, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 1.95432388517783, | |
| "grad_norm": 17.767173767089844, | |
| "learning_rate": 7.612838515546641e-07, | |
| "loss": 0.4525, | |
| "step": 21650 | |
| }, | |
| { | |
| "epoch": 1.958837335259072, | |
| "grad_norm": 15.962186813354492, | |
| "learning_rate": 6.860581745235707e-07, | |
| "loss": 0.6349, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 1.9633507853403143, | |
| "grad_norm": 3.5042107105255127, | |
| "learning_rate": 6.108324974924774e-07, | |
| "loss": 0.4903, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 1.9633507853403143, | |
| "eval_exact_match": 86.30085146641439, | |
| "eval_f1": 92.62412689139829, | |
| "eval_runtime": 143.4548, | |
| "eval_samples_per_second": 75.271, | |
| "eval_steps_per_second": 18.821, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 1.967864235421556, | |
| "grad_norm": 3.967465877532959, | |
| "learning_rate": 5.356068204613842e-07, | |
| "loss": 0.4477, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 1.9723776855027984, | |
| "grad_norm": 13.418035507202148, | |
| "learning_rate": 4.603811434302909e-07, | |
| "loss": 0.4418, | |
| "step": 21850 | |
| }, | |
| { | |
| "epoch": 1.9768911355840404, | |
| "grad_norm": 8.183111190795898, | |
| "learning_rate": 3.851554663991976e-07, | |
| "loss": 0.5417, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 1.9814045856652824, | |
| "grad_norm": 5.646338939666748, | |
| "learning_rate": 3.099297893681043e-07, | |
| "loss": 0.5065, | |
| "step": 21950 | |
| }, | |
| { | |
| "epoch": 1.9859180357465247, | |
| "grad_norm": 14.187732696533203, | |
| "learning_rate": 2.3470411233701103e-07, | |
| "loss": 0.4325, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.9859180357465247, | |
| "eval_exact_match": 86.35761589403974, | |
| "eval_f1": 92.66264597808306, | |
| "eval_runtime": 143.145, | |
| "eval_samples_per_second": 75.434, | |
| "eval_steps_per_second": 18.862, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.9904314858277667, | |
| "grad_norm": 24.069639205932617, | |
| "learning_rate": 1.5947843530591774e-07, | |
| "loss": 0.4079, | |
| "step": 22050 | |
| }, | |
| { | |
| "epoch": 1.9949449359090088, | |
| "grad_norm": 9.552345275878906, | |
| "learning_rate": 8.425275827482447e-08, | |
| "loss": 0.4418, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 1.999458385990251, | |
| "grad_norm": 11.15715503692627, | |
| "learning_rate": 9.027081243731194e-09, | |
| "loss": 0.519, | |
| "step": 22150 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 22156, | |
| "total_flos": 4.529540706059981e+16, | |
| "train_loss": 0.8456309766066937, | |
| "train_runtime": 23960.3876, | |
| "train_samples_per_second": 7.397, | |
| "train_steps_per_second": 0.925 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 22156, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 4.529540706059981e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |