| { | |
| "best_global_step": 5822, | |
| "best_metric": 1.5156257152557373, | |
| "best_model_checkpoint": "./output/bert-base-medmentions/checkpoint-5822", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 43665, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1717622810030917, | |
| "grad_norm": 1.9404675960540771, | |
| "learning_rate": 4.9427459063323026e-05, | |
| "loss": 1.7239, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3435245620061834, | |
| "grad_norm": 1.6719824075698853, | |
| "learning_rate": 4.885491812664606e-05, | |
| "loss": 1.6163, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5152868430092752, | |
| "grad_norm": 1.837388515472412, | |
| "learning_rate": 4.828237718996909e-05, | |
| "loss": 1.5924, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6870491240123668, | |
| "grad_norm": 2.1180434226989746, | |
| "learning_rate": 4.770983625329211e-05, | |
| "loss": 1.5717, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8588114050154586, | |
| "grad_norm": 1.800995945930481, | |
| "learning_rate": 4.7137295316615135e-05, | |
| "loss": 1.5686, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8549719237162791, | |
| "eval_f1": 0.5650458863307036, | |
| "eval_loss": 1.5439889430999756, | |
| "eval_precision": 0.5245889387144993, | |
| "eval_recall": 0.6122644801116539, | |
| "eval_runtime": 4.1137, | |
| "eval_samples_per_second": 707.39, | |
| "eval_steps_per_second": 88.485, | |
| "step": 2911 | |
| }, | |
| { | |
| "epoch": 1.0305736860185504, | |
| "grad_norm": 1.0917384624481201, | |
| "learning_rate": 4.6564754379938166e-05, | |
| "loss": 1.5484, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.202335967021642, | |
| "grad_norm": 1.0721220970153809, | |
| "learning_rate": 4.59922134432612e-05, | |
| "loss": 1.4824, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3740982480247337, | |
| "grad_norm": 1.5045437812805176, | |
| "learning_rate": 4.541967250658422e-05, | |
| "loss": 1.4894, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5458605290278253, | |
| "grad_norm": 3.3368799686431885, | |
| "learning_rate": 4.484713156990725e-05, | |
| "loss": 1.4863, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.7176228100309172, | |
| "grad_norm": 2.191357374191284, | |
| "learning_rate": 4.427459063323028e-05, | |
| "loss": 1.4806, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.889385091034009, | |
| "grad_norm": 2.575375556945801, | |
| "learning_rate": 4.3702049696553306e-05, | |
| "loss": 1.4792, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8688595400463357, | |
| "eval_f1": 0.6071204975165909, | |
| "eval_loss": 1.5156257152557373, | |
| "eval_precision": 0.5820728291316527, | |
| "eval_recall": 0.6344207955338451, | |
| "eval_runtime": 4.3513, | |
| "eval_samples_per_second": 668.77, | |
| "eval_steps_per_second": 83.654, | |
| "step": 5822 | |
| }, | |
| { | |
| "epoch": 2.0611473720371007, | |
| "grad_norm": 4.3115081787109375, | |
| "learning_rate": 4.3129508759876336e-05, | |
| "loss": 1.4566, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.2329096530401924, | |
| "grad_norm": 1.5031124353408813, | |
| "learning_rate": 4.255696782319936e-05, | |
| "loss": 1.4053, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.404671934043284, | |
| "grad_norm": 2.4502875804901123, | |
| "learning_rate": 4.198442688652239e-05, | |
| "loss": 1.4111, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.5764342150463757, | |
| "grad_norm": 1.5572278499603271, | |
| "learning_rate": 4.1411885949845415e-05, | |
| "loss": 1.4096, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.7481964960494674, | |
| "grad_norm": 1.3663930892944336, | |
| "learning_rate": 4.0839345013168445e-05, | |
| "loss": 1.4097, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.9199587770525595, | |
| "grad_norm": 3.3840882778167725, | |
| "learning_rate": 4.026680407649147e-05, | |
| "loss": 1.4111, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8713595727692771, | |
| "eval_f1": 0.6163334575106585, | |
| "eval_loss": 1.519059419631958, | |
| "eval_precision": 0.5864513588026782, | |
| "eval_recall": 0.6494242847173761, | |
| "eval_runtime": 4.33, | |
| "eval_samples_per_second": 672.06, | |
| "eval_steps_per_second": 84.065, | |
| "step": 8733 | |
| }, | |
| { | |
| "epoch": 3.091721058055651, | |
| "grad_norm": 1.4997501373291016, | |
| "learning_rate": 3.96942631398145e-05, | |
| "loss": 1.3798, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.2634833390587428, | |
| "grad_norm": 1.4717656373977661, | |
| "learning_rate": 3.912172220313753e-05, | |
| "loss": 1.3517, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.4352456200618344, | |
| "grad_norm": 2.0151214599609375, | |
| "learning_rate": 3.8549181266460554e-05, | |
| "loss": 1.3532, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.607007901064926, | |
| "grad_norm": 2.7060940265655518, | |
| "learning_rate": 3.797664032978358e-05, | |
| "loss": 1.3546, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.7787701820680177, | |
| "grad_norm": 3.2001101970672607, | |
| "learning_rate": 3.740409939310661e-05, | |
| "loss": 1.355, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.9505324630711094, | |
| "grad_norm": 2.374141216278076, | |
| "learning_rate": 3.683155845642964e-05, | |
| "loss": 1.356, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8777209125773897, | |
| "eval_f1": 0.6318054658919734, | |
| "eval_loss": 1.5293220281600952, | |
| "eval_precision": 0.6235663919802905, | |
| "eval_recall": 0.6402651779483601, | |
| "eval_runtime": 4.3262, | |
| "eval_samples_per_second": 672.648, | |
| "eval_steps_per_second": 84.139, | |
| "step": 11644 | |
| }, | |
| { | |
| "epoch": 4.1222947440742015, | |
| "grad_norm": 2.5122597217559814, | |
| "learning_rate": 3.6259017519752663e-05, | |
| "loss": 1.3298, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.294057025077293, | |
| "grad_norm": 1.870731234550476, | |
| "learning_rate": 3.568647658307569e-05, | |
| "loss": 1.3148, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.465819306080385, | |
| "grad_norm": 2.1130025386810303, | |
| "learning_rate": 3.511393564639872e-05, | |
| "loss": 1.3147, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.637581587083476, | |
| "grad_norm": 1.9891668558120728, | |
| "learning_rate": 3.454139470972175e-05, | |
| "loss": 1.3176, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.809343868086568, | |
| "grad_norm": 2.144550085067749, | |
| "learning_rate": 3.396885377304477e-05, | |
| "loss": 1.3207, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.98110614908966, | |
| "grad_norm": 1.166013240814209, | |
| "learning_rate": 3.33963128363678e-05, | |
| "loss": 1.3182, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8789120276443408, | |
| "eval_f1": 0.6353600689952565, | |
| "eval_loss": 1.543265461921692, | |
| "eval_precision": 0.6282619819205185, | |
| "eval_recall": 0.6426203768318214, | |
| "eval_runtime": 4.3312, | |
| "eval_samples_per_second": 671.862, | |
| "eval_steps_per_second": 84.041, | |
| "step": 14555 | |
| }, | |
| { | |
| "epoch": 5.152868430092751, | |
| "grad_norm": 1.4789066314697266, | |
| "learning_rate": 3.2823771899690834e-05, | |
| "loss": 1.2897, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.3246307110958435, | |
| "grad_norm": 2.067422866821289, | |
| "learning_rate": 3.225123096301386e-05, | |
| "loss": 1.2908, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.496392992098935, | |
| "grad_norm": 1.4258556365966797, | |
| "learning_rate": 3.167869002633688e-05, | |
| "loss": 1.2946, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.668155273102027, | |
| "grad_norm": 1.9385099411010742, | |
| "learning_rate": 3.110614908965991e-05, | |
| "loss": 1.2914, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.839917554105119, | |
| "grad_norm": 2.5622832775115967, | |
| "learning_rate": 3.053360815298294e-05, | |
| "loss": 1.2919, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8793701488239375, | |
| "eval_f1": 0.6428903837204383, | |
| "eval_loss": 1.5671014785766602, | |
| "eval_precision": 0.6241682411895177, | |
| "eval_recall": 0.6627704117236567, | |
| "eval_runtime": 4.523, | |
| "eval_samples_per_second": 643.376, | |
| "eval_steps_per_second": 80.477, | |
| "step": 17466 | |
| }, | |
| { | |
| "epoch": 6.01167983510821, | |
| "grad_norm": 1.9045183658599854, | |
| "learning_rate": 2.9961067216305967e-05, | |
| "loss": 1.2899, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 6.183442116111302, | |
| "grad_norm": 5.356103420257568, | |
| "learning_rate": 2.9388526279628997e-05, | |
| "loss": 1.2716, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.3552043971143934, | |
| "grad_norm": 1.4505314826965332, | |
| "learning_rate": 2.8815985342952025e-05, | |
| "loss": 1.2736, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.5269666781174855, | |
| "grad_norm": 2.0673441886901855, | |
| "learning_rate": 2.824344440627505e-05, | |
| "loss": 1.2755, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.698728959120577, | |
| "grad_norm": 7.07130765914917, | |
| "learning_rate": 2.7670903469598076e-05, | |
| "loss": 1.2739, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.870491240123669, | |
| "grad_norm": 2.692305326461792, | |
| "learning_rate": 2.7098362532921106e-05, | |
| "loss": 1.2743, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8809277608345659, | |
| "eval_f1": 0.6462844646455984, | |
| "eval_loss": 1.5696512460708618, | |
| "eval_precision": 0.6355739225773804, | |
| "eval_recall": 0.6573621772505234, | |
| "eval_runtime": 4.3283, | |
| "eval_samples_per_second": 672.312, | |
| "eval_steps_per_second": 84.097, | |
| "step": 20377 | |
| }, | |
| { | |
| "epoch": 7.042253521126761, | |
| "grad_norm": 0.9224966764450073, | |
| "learning_rate": 2.6525821596244134e-05, | |
| "loss": 1.2716, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 7.214015802129852, | |
| "grad_norm": 2.72609543800354, | |
| "learning_rate": 2.5953280659567158e-05, | |
| "loss": 1.2578, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 7.385778083132944, | |
| "grad_norm": 1.7019892930984497, | |
| "learning_rate": 2.538073972289019e-05, | |
| "loss": 1.2612, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 7.5575403641360355, | |
| "grad_norm": 4.017130374908447, | |
| "learning_rate": 2.4808198786213216e-05, | |
| "loss": 1.2622, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.729302645139128, | |
| "grad_norm": 6.522401332855225, | |
| "learning_rate": 2.4235657849536243e-05, | |
| "loss": 1.2611, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.901064926142219, | |
| "grad_norm": 3.355700731277466, | |
| "learning_rate": 2.366311691285927e-05, | |
| "loss": 1.2633, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8812811686016833, | |
| "eval_f1": 0.6527559389741191, | |
| "eval_loss": 1.5806214809417725, | |
| "eval_precision": 0.636446507002569, | |
| "eval_recall": 0.6699232379623168, | |
| "eval_runtime": 4.329, | |
| "eval_samples_per_second": 672.206, | |
| "eval_steps_per_second": 84.084, | |
| "step": 23288 | |
| }, | |
| { | |
| "epoch": 8.07282720714531, | |
| "grad_norm": 0.5228517651557922, | |
| "learning_rate": 2.30905759761823e-05, | |
| "loss": 1.2579, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 8.244589488148403, | |
| "grad_norm": 0.2985314726829529, | |
| "learning_rate": 2.2518035039505325e-05, | |
| "loss": 1.2527, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 8.416351769151495, | |
| "grad_norm": 1.963086724281311, | |
| "learning_rate": 2.1945494102828355e-05, | |
| "loss": 1.2542, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 8.588114050154585, | |
| "grad_norm": 0.8812742233276367, | |
| "learning_rate": 2.137295316615138e-05, | |
| "loss": 1.251, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.759876331157677, | |
| "grad_norm": 0.7020455002784729, | |
| "learning_rate": 2.080041222947441e-05, | |
| "loss": 1.2516, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 8.93163861216077, | |
| "grad_norm": 2.0791664123535156, | |
| "learning_rate": 2.0227871292797437e-05, | |
| "loss": 1.2542, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.880783779892407, | |
| "eval_f1": 0.6498043011657758, | |
| "eval_loss": 1.594204068183899, | |
| "eval_precision": 0.6277953972513621, | |
| "eval_recall": 0.6734124214933705, | |
| "eval_runtime": 4.3509, | |
| "eval_samples_per_second": 668.824, | |
| "eval_steps_per_second": 83.66, | |
| "step": 26199 | |
| }, | |
| { | |
| "epoch": 9.103400893163862, | |
| "grad_norm": 2.5551717281341553, | |
| "learning_rate": 1.9655330356120464e-05, | |
| "loss": 1.2496, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 9.275163174166954, | |
| "grad_norm": 3.963749647140503, | |
| "learning_rate": 1.908278941944349e-05, | |
| "loss": 1.2453, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 9.446925455170044, | |
| "grad_norm": 0.5859785676002502, | |
| "learning_rate": 1.851024848276652e-05, | |
| "loss": 1.2463, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 9.618687736173136, | |
| "grad_norm": 0.3447531759738922, | |
| "learning_rate": 1.7937707546089546e-05, | |
| "loss": 1.2477, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 9.790450017176228, | |
| "grad_norm": 0.3794388175010681, | |
| "learning_rate": 1.7365166609412573e-05, | |
| "loss": 1.2468, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 9.96221229817932, | |
| "grad_norm": 1.6076109409332275, | |
| "learning_rate": 1.67926256727356e-05, | |
| "loss": 1.2457, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8814251495438422, | |
| "eval_f1": 0.6500277789649131, | |
| "eval_loss": 1.607577919960022, | |
| "eval_precision": 0.6372015081692501, | |
| "eval_recall": 0.6633810188415911, | |
| "eval_runtime": 4.3517, | |
| "eval_samples_per_second": 668.699, | |
| "eval_steps_per_second": 83.645, | |
| "step": 29110 | |
| }, | |
| { | |
| "epoch": 10.13397457918241, | |
| "grad_norm": 1.843865990638733, | |
| "learning_rate": 1.622008473605863e-05, | |
| "loss": 1.2411, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 10.305736860185503, | |
| "grad_norm": 2.924731731414795, | |
| "learning_rate": 1.5647543799381655e-05, | |
| "loss": 1.2402, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 10.477499141188595, | |
| "grad_norm": 0.40096113085746765, | |
| "learning_rate": 1.5075002862704684e-05, | |
| "loss": 1.2416, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 10.649261422191687, | |
| "grad_norm": 1.3067082166671753, | |
| "learning_rate": 1.4502461926027711e-05, | |
| "loss": 1.2422, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 10.82102370319478, | |
| "grad_norm": 1.1540168523788452, | |
| "learning_rate": 1.3929920989350739e-05, | |
| "loss": 1.2406, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 10.99278598419787, | |
| "grad_norm": 2.354355812072754, | |
| "learning_rate": 1.3357380052673768e-05, | |
| "loss": 1.2398, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8834932394403068, | |
| "eval_f1": 0.6551724137931034, | |
| "eval_loss": 1.6077239513397217, | |
| "eval_precision": 0.6413770053475936, | |
| "eval_recall": 0.6695743196092114, | |
| "eval_runtime": 4.5309, | |
| "eval_samples_per_second": 642.254, | |
| "eval_steps_per_second": 80.337, | |
| "step": 32021 | |
| }, | |
| { | |
| "epoch": 11.164548265200962, | |
| "grad_norm": 2.941948413848877, | |
| "learning_rate": 1.2784839115996793e-05, | |
| "loss": 1.2384, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 11.336310546204054, | |
| "grad_norm": 0.06978488713502884, | |
| "learning_rate": 1.2212298179319822e-05, | |
| "loss": 1.238, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 11.508072827207146, | |
| "grad_norm": 1.1202852725982666, | |
| "learning_rate": 1.163975724264285e-05, | |
| "loss": 1.2375, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 11.679835108210238, | |
| "grad_norm": 3.7290749549865723, | |
| "learning_rate": 1.1067216305965877e-05, | |
| "loss": 1.2373, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 11.851597389213328, | |
| "grad_norm": 0.5790780782699585, | |
| "learning_rate": 1.0494675369288904e-05, | |
| "loss": 1.2377, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8846974436838179, | |
| "eval_f1": 0.6615437158469945, | |
| "eval_loss": 1.6134886741638184, | |
| "eval_precision": 0.6478260869565218, | |
| "eval_recall": 0.6758548499651081, | |
| "eval_runtime": 4.3276, | |
| "eval_samples_per_second": 672.43, | |
| "eval_steps_per_second": 84.112, | |
| "step": 34932 | |
| }, | |
| { | |
| "epoch": 12.02335967021642, | |
| "grad_norm": 0.04744827747344971, | |
| "learning_rate": 9.922134432611933e-06, | |
| "loss": 1.2362, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 12.195121951219512, | |
| "grad_norm": 0.07846707850694656, | |
| "learning_rate": 9.34959349593496e-06, | |
| "loss": 1.235, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 12.366884232222604, | |
| "grad_norm": 3.9505062103271484, | |
| "learning_rate": 8.777052559257987e-06, | |
| "loss": 1.2345, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 12.538646513225697, | |
| "grad_norm": 0.13419800996780396, | |
| "learning_rate": 8.204511622581015e-06, | |
| "loss": 1.2362, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 12.710408794228787, | |
| "grad_norm": 0.205936998128891, | |
| "learning_rate": 7.631970685904042e-06, | |
| "loss": 1.2341, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 12.882171075231879, | |
| "grad_norm": 1.917006254196167, | |
| "learning_rate": 7.05942974922707e-06, | |
| "loss": 1.2349, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.883872825560544, | |
| "eval_f1": 0.6590367597004765, | |
| "eval_loss": 1.619519829750061, | |
| "eval_precision": 0.6432724252491694, | |
| "eval_recall": 0.6755931612002791, | |
| "eval_runtime": 4.3511, | |
| "eval_samples_per_second": 668.801, | |
| "eval_steps_per_second": 83.658, | |
| "step": 37843 | |
| }, | |
| { | |
| "epoch": 13.053933356234971, | |
| "grad_norm": 0.5815674662590027, | |
| "learning_rate": 6.486888812550097e-06, | |
| "loss": 1.2342, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 13.225695637238063, | |
| "grad_norm": 0.48151713609695435, | |
| "learning_rate": 5.914347875873125e-06, | |
| "loss": 1.2335, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 13.397457918241154, | |
| "grad_norm": 4.141974925994873, | |
| "learning_rate": 5.341806939196153e-06, | |
| "loss": 1.2335, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 13.569220199244246, | |
| "grad_norm": 0.24046790599822998, | |
| "learning_rate": 4.76926600251918e-06, | |
| "loss": 1.2331, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 13.740982480247338, | |
| "grad_norm": 0.08363146334886551, | |
| "learning_rate": 4.196725065842208e-06, | |
| "loss": 1.233, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 13.91274476125043, | |
| "grad_norm": 0.5658828616142273, | |
| "learning_rate": 3.6241841291652353e-06, | |
| "loss": 1.2328, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8845272843885391, | |
| "eval_f1": 0.6591725081210464, | |
| "eval_loss": 1.6228290796279907, | |
| "eval_precision": 0.6462453905464298, | |
| "eval_recall": 0.6726273551988835, | |
| "eval_runtime": 4.3411, | |
| "eval_samples_per_second": 670.343, | |
| "eval_steps_per_second": 83.85, | |
| "step": 40754 | |
| }, | |
| { | |
| "epoch": 14.084507042253522, | |
| "grad_norm": 1.4916341304779053, | |
| "learning_rate": 3.051643192488263e-06, | |
| "loss": 1.2318, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 14.256269323256612, | |
| "grad_norm": 0.1434667557477951, | |
| "learning_rate": 2.4791022558112906e-06, | |
| "loss": 1.2305, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 14.428031604259704, | |
| "grad_norm": 0.10652283579111099, | |
| "learning_rate": 1.906561319134318e-06, | |
| "loss": 1.232, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 14.599793885262796, | |
| "grad_norm": 0.8040905594825745, | |
| "learning_rate": 1.3340203824573458e-06, | |
| "loss": 1.2321, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 14.771556166265889, | |
| "grad_norm": 0.040788378566503525, | |
| "learning_rate": 7.614794457803733e-07, | |
| "loss": 1.2319, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 14.943318447268979, | |
| "grad_norm": 0.5179036259651184, | |
| "learning_rate": 1.889385091034009e-07, | |
| "loss": 1.231, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847236220369377, | |
| "eval_f1": 0.660140218878249, | |
| "eval_loss": 1.6247130632400513, | |
| "eval_precision": 0.6473004694835681, | |
| "eval_recall": 0.6734996510816469, | |
| "eval_runtime": 4.3445, | |
| "eval_samples_per_second": 669.806, | |
| "eval_steps_per_second": 83.783, | |
| "step": 43665 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 43665, | |
| "total_flos": 1.1901430945516224e+16, | |
| "train_loss": 1.311661433704009, | |
| "train_runtime": 2299.911, | |
| "train_samples_per_second": 151.865, | |
| "train_steps_per_second": 18.986 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 43665, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 1.1901430945516224e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |