| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.923076923076923, |
| "eval_steps": 500, |
| "global_step": 462, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0641025641025641, |
| "grad_norm": 3.655071258544922, |
| "learning_rate": 1.3636363636363635e-05, |
| "loss": 1.8656, |
| "mean_token_accuracy": 0.6241337656974792, |
| "num_tokens": 607410.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1282051282051282, |
| "grad_norm": 1.5179063081741333, |
| "learning_rate": 3.068181818181818e-05, |
| "loss": 1.6667, |
| "mean_token_accuracy": 0.6309765815734864, |
| "num_tokens": 1203642.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.19230769230769232, |
| "grad_norm": 0.7223508358001709, |
| "learning_rate": 4.7727272727272724e-05, |
| "loss": 1.3409, |
| "mean_token_accuracy": 0.6678204417228699, |
| "num_tokens": 1801470.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2564102564102564, |
| "grad_norm": 0.6648445129394531, |
| "learning_rate": 6.477272727272727e-05, |
| "loss": 1.1631, |
| "mean_token_accuracy": 0.6982533812522889, |
| "num_tokens": 2372190.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.32051282051282054, |
| "grad_norm": 0.4945732355117798, |
| "learning_rate": 7.499720909295504e-05, |
| "loss": 1.0379, |
| "mean_token_accuracy": 0.7267868638038635, |
| "num_tokens": 2961891.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.38461538461538464, |
| "grad_norm": 0.5823026299476624, |
| "learning_rate": 7.496581774950121e-05, |
| "loss": 0.9857, |
| "mean_token_accuracy": 0.7317686080932617, |
| "num_tokens": 3548150.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.44871794871794873, |
| "grad_norm": 0.41620248556137085, |
| "learning_rate": 7.48995854931481e-05, |
| "loss": 0.8981, |
| "mean_token_accuracy": 0.7510427117347718, |
| "num_tokens": 4203006.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.5128205128205128, |
| "grad_norm": 0.5902088284492493, |
| "learning_rate": 7.47985944714606e-05, |
| "loss": 0.8346, |
| "mean_token_accuracy": 0.7676318407058715, |
| "num_tokens": 4759630.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5769230769230769, |
| "grad_norm": 0.45137742161750793, |
| "learning_rate": 7.466296994314145e-05, |
| "loss": 0.7912, |
| "mean_token_accuracy": 0.7737884044647216, |
| "num_tokens": 5355141.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.6410256410256411, |
| "grad_norm": 0.4960101842880249, |
| "learning_rate": 7.449288012267344e-05, |
| "loss": 0.7712, |
| "mean_token_accuracy": 0.7818852543830872, |
| "num_tokens": 5922140.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.7051282051282052, |
| "grad_norm": 0.5259331464767456, |
| "learning_rate": 7.428853597168368e-05, |
| "loss": 0.7453, |
| "mean_token_accuracy": 0.7817357063293457, |
| "num_tokens": 6555424.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.7692307692307693, |
| "grad_norm": 0.5303130745887756, |
| "learning_rate": 7.405019093728897e-05, |
| "loss": 0.6768, |
| "mean_token_accuracy": 0.8006563067436219, |
| "num_tokens": 7160870.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 0.47481822967529297, |
| "learning_rate": 7.377814063774636e-05, |
| "loss": 0.7071, |
| "mean_token_accuracy": 0.7933502435684204, |
| "num_tokens": 7760564.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.8974358974358975, |
| "grad_norm": 0.4843287169933319, |
| "learning_rate": 7.34727224957993e-05, |
| "loss": 0.683, |
| "mean_token_accuracy": 0.8000761032104492, |
| "num_tokens": 8370438.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.9615384615384616, |
| "grad_norm": 0.5345025658607483, |
| "learning_rate": 7.313431532017372e-05, |
| "loss": 0.6762, |
| "mean_token_accuracy": 0.7997780799865722, |
| "num_tokens": 8968243.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.0256410256410255, |
| "grad_norm": 0.5506055355072021, |
| "learning_rate": 7.276333883574324e-05, |
| "loss": 0.6194, |
| "mean_token_accuracy": 0.8143499016761779, |
| "num_tokens": 9582180.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.0897435897435896, |
| "grad_norm": 0.5209991931915283, |
| "learning_rate": 7.236025316294642e-05, |
| "loss": 0.609, |
| "mean_token_accuracy": 0.8159718990325928, |
| "num_tokens": 10172288.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.1538461538461537, |
| "grad_norm": 0.6436333656311035, |
| "learning_rate": 7.192555824710139e-05, |
| "loss": 0.6107, |
| "mean_token_accuracy": 0.8144642829895019, |
| "num_tokens": 10769484.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.217948717948718, |
| "grad_norm": 0.5410007834434509, |
| "learning_rate": 7.145979323832603e-05, |
| "loss": 0.6005, |
| "mean_token_accuracy": 0.8184427499771119, |
| "num_tokens": 11375057.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.282051282051282, |
| "grad_norm": 0.6425564885139465, |
| "learning_rate": 7.096353582283249e-05, |
| "loss": 0.5923, |
| "mean_token_accuracy": 0.822891366481781, |
| "num_tokens": 11922877.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.3461538461538463, |
| "grad_norm": 0.5346367955207825, |
| "learning_rate": 7.043740150642567e-05, |
| "loss": 0.5725, |
| "mean_token_accuracy": 0.8216536045074463, |
| "num_tokens": 12535517.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.4102564102564101, |
| "grad_norm": 0.7521743178367615, |
| "learning_rate": 6.988204285109415e-05, |
| "loss": 0.5484, |
| "mean_token_accuracy": 0.82740398645401, |
| "num_tokens": 13117170.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.4743589743589745, |
| "grad_norm": 0.5057587027549744, |
| "learning_rate": 6.929814866564053e-05, |
| "loss": 0.569, |
| "mean_token_accuracy": 0.8290618062019348, |
| "num_tokens": 13732822.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.5384615384615383, |
| "grad_norm": 0.5119039416313171, |
| "learning_rate": 6.868644315135503e-05, |
| "loss": 0.5928, |
| "mean_token_accuracy": 0.8159135937690735, |
| "num_tokens": 14352493.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.6025641025641026, |
| "grad_norm": 0.6582021713256836, |
| "learning_rate": 6.804768500379196e-05, |
| "loss": 0.5818, |
| "mean_token_accuracy": 0.8214109659194946, |
| "num_tokens": 14977100.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.5718688368797302, |
| "learning_rate": 6.738266647176307e-05, |
| "loss": 0.5356, |
| "mean_token_accuracy": 0.8339002370834351, |
| "num_tokens": 15579948.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.7307692307692308, |
| "grad_norm": 0.708949089050293, |
| "learning_rate": 6.669221237471496e-05, |
| "loss": 0.5682, |
| "mean_token_accuracy": 0.8245456099510193, |
| "num_tokens": 16171972.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.7948717948717947, |
| "grad_norm": 0.6917956471443176, |
| "learning_rate": 6.597717907970928e-05, |
| "loss": 0.5634, |
| "mean_token_accuracy": 0.8291388511657715, |
| "num_tokens": 16740529.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.858974358974359, |
| "grad_norm": 0.5983976125717163, |
| "learning_rate": 6.523845343927467e-05, |
| "loss": 0.5589, |
| "mean_token_accuracy": 0.827668035030365, |
| "num_tokens": 17360396.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.9230769230769231, |
| "grad_norm": 0.6851189732551575, |
| "learning_rate": 6.447695169144747e-05, |
| "loss": 0.5202, |
| "mean_token_accuracy": 0.8363027453422547, |
| "num_tokens": 17950794.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.9871794871794872, |
| "grad_norm": 0.6370279788970947, |
| "learning_rate": 6.3693618323366e-05, |
| "loss": 0.5438, |
| "mean_token_accuracy": 0.8309719562530518, |
| "num_tokens": 18556813.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.051282051282051, |
| "grad_norm": 0.653179407119751, |
| "learning_rate": 6.288942489982737e-05, |
| "loss": 0.5018, |
| "mean_token_accuracy": 0.8419281959533691, |
| "num_tokens": 19156646.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.1153846153846154, |
| "grad_norm": 0.7452319264411926, |
| "learning_rate": 6.206536885826024e-05, |
| "loss": 0.4707, |
| "mean_token_accuracy": 0.849298357963562, |
| "num_tokens": 19728876.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.1794871794871793, |
| "grad_norm": 0.6136786341667175, |
| "learning_rate": 6.122247227160757e-05, |
| "loss": 0.4443, |
| "mean_token_accuracy": 0.8596317291259765, |
| "num_tokens": 20296852.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.2435897435897436, |
| "grad_norm": 0.7459028959274292, |
| "learning_rate": 6.0361780580654354e-05, |
| "loss": 0.5131, |
| "mean_token_accuracy": 0.841215193271637, |
| "num_tokens": 20910521.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.3076923076923075, |
| "grad_norm": 0.6735811829566956, |
| "learning_rate": 5.94843612973721e-05, |
| "loss": 0.4646, |
| "mean_token_accuracy": 0.8512268900871277, |
| "num_tokens": 21522235.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.371794871794872, |
| "grad_norm": 0.7015240788459778, |
| "learning_rate": 5.8591302680888696e-05, |
| "loss": 0.4931, |
| "mean_token_accuracy": 0.8425710797309875, |
| "num_tokens": 22149827.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.435897435897436, |
| "grad_norm": 0.6728143692016602, |
| "learning_rate": 5.76837123877255e-05, |
| "loss": 0.4617, |
| "mean_token_accuracy": 0.8551092624664307, |
| "num_tokens": 22729797.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.799096941947937, |
| "learning_rate": 5.6762716097976086e-05, |
| "loss": 0.462, |
| "mean_token_accuracy": 0.8515766501426697, |
| "num_tokens": 23337181.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.564102564102564, |
| "grad_norm": 0.7390243411064148, |
| "learning_rate": 5.582945611913029e-05, |
| "loss": 0.4642, |
| "mean_token_accuracy": 0.8526711583137512, |
| "num_tokens": 23912861.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.628205128205128, |
| "grad_norm": 0.6593577861785889, |
| "learning_rate": 5.4885089969275554e-05, |
| "loss": 0.4615, |
| "mean_token_accuracy": 0.854354739189148, |
| "num_tokens": 24501149.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.6923076923076925, |
| "grad_norm": 0.7466415166854858, |
| "learning_rate": 5.393078894143243e-05, |
| "loss": 0.482, |
| "mean_token_accuracy": 0.8460716485977173, |
| "num_tokens": 25121188.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.7564102564102564, |
| "grad_norm": 0.8499466776847839, |
| "learning_rate": 5.296773665080527e-05, |
| "loss": 0.486, |
| "mean_token_accuracy": 0.8459465742111206, |
| "num_tokens": 25716660.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.8205128205128203, |
| "grad_norm": 0.6929744482040405, |
| "learning_rate": 5.1997127566749675e-05, |
| "loss": 0.4631, |
| "mean_token_accuracy": 0.8534524321556092, |
| "num_tokens": 26349046.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.8846153846153846, |
| "grad_norm": 0.7020038366317749, |
| "learning_rate": 5.102016553127759e-05, |
| "loss": 0.4577, |
| "mean_token_accuracy": 0.8514141321182251, |
| "num_tokens": 26932820.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.948717948717949, |
| "grad_norm": 0.8280526399612427, |
| "learning_rate": 5.0038062265937555e-05, |
| "loss": 0.4619, |
| "mean_token_accuracy": 0.8519901514053345, |
| "num_tokens": 27533753.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 3.0128205128205128, |
| "grad_norm": 0.7628069519996643, |
| "learning_rate": 4.9052035868922074e-05, |
| "loss": 0.4373, |
| "mean_token_accuracy": 0.8593450307846069, |
| "num_tokens": 28120086.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 3.076923076923077, |
| "grad_norm": 0.7876071929931641, |
| "learning_rate": 4.80633093042659e-05, |
| "loss": 0.4109, |
| "mean_token_accuracy": 0.8663130760192871, |
| "num_tokens": 28722867.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 3.141025641025641, |
| "grad_norm": 0.8466971516609192, |
| "learning_rate": 4.7073108885009375e-05, |
| "loss": 0.3831, |
| "mean_token_accuracy": 0.8726190209388733, |
| "num_tokens": 29295716.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 3.2051282051282053, |
| "grad_norm": 0.7627570629119873, |
| "learning_rate": 4.6082662752208e-05, |
| "loss": 0.3759, |
| "mean_token_accuracy": 0.8777644276618958, |
| "num_tokens": 29897685.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 3.269230769230769, |
| "grad_norm": 0.7697687745094299, |
| "learning_rate": 4.509319935167465e-05, |
| "loss": 0.3899, |
| "mean_token_accuracy": 0.871421754360199, |
| "num_tokens": 30488009.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.9552246928215027, |
| "learning_rate": 4.410594591034382e-05, |
| "loss": 0.383, |
| "mean_token_accuracy": 0.8755820751190185, |
| "num_tokens": 31058890.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 3.3974358974358974, |
| "grad_norm": 0.8885680437088013, |
| "learning_rate": 4.312212691414775e-05, |
| "loss": 0.4121, |
| "mean_token_accuracy": 0.8662847876548767, |
| "num_tokens": 31662307.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.4615384615384617, |
| "grad_norm": 0.7865875363349915, |
| "learning_rate": 4.2142962589292045e-05, |
| "loss": 0.3771, |
| "mean_token_accuracy": 0.8753103733062744, |
| "num_tokens": 32245760.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.5256410256410255, |
| "grad_norm": 0.7509394288063049, |
| "learning_rate": 4.1169667388814757e-05, |
| "loss": 0.4164, |
| "mean_token_accuracy": 0.862480902671814, |
| "num_tokens": 32843525.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.58974358974359, |
| "grad_norm": 1.0797700881958008, |
| "learning_rate": 4.0203448486305725e-05, |
| "loss": 0.4084, |
| "mean_token_accuracy": 0.8689486265182496, |
| "num_tokens": 33471785.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.6538461538461537, |
| "grad_norm": 0.8257498741149902, |
| "learning_rate": 3.92455042786548e-05, |
| "loss": 0.4062, |
| "mean_token_accuracy": 0.8635867476463318, |
| "num_tokens": 34104891.0, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.717948717948718, |
| "grad_norm": 0.8262436389923096, |
| "learning_rate": 3.8297022899685596e-05, |
| "loss": 0.4139, |
| "mean_token_accuracy": 0.8653790712356567, |
| "num_tokens": 34698062.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.782051282051282, |
| "grad_norm": 0.7923848628997803, |
| "learning_rate": 3.735918074651855e-05, |
| "loss": 0.4179, |
| "mean_token_accuracy": 0.8631722807884217, |
| "num_tokens": 35319450.0, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.8461538461538463, |
| "grad_norm": 0.6863822937011719, |
| "learning_rate": 3.643314102049095e-05, |
| "loss": 0.4058, |
| "mean_token_accuracy": 0.8680677890777588, |
| "num_tokens": 35905988.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.91025641025641, |
| "grad_norm": 0.9361572265625, |
| "learning_rate": 3.552005228444363e-05, |
| "loss": 0.4012, |
| "mean_token_accuracy": 0.8704957604408264, |
| "num_tokens": 36458945.0, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.9743589743589745, |
| "grad_norm": 0.7748892903327942, |
| "learning_rate": 3.462104703816375e-05, |
| "loss": 0.3719, |
| "mean_token_accuracy": 0.8755459785461426, |
| "num_tokens": 37081595.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 4.038461538461538, |
| "grad_norm": 0.7349646091461182, |
| "learning_rate": 3.3737240313750425e-05, |
| "loss": 0.3389, |
| "mean_token_accuracy": 0.8895657777786254, |
| "num_tokens": 37686723.0, |
| "step": 315 |
| }, |
| { |
| "epoch": 4.102564102564102, |
| "grad_norm": 0.8582376837730408, |
| "learning_rate": 3.286972829264558e-05, |
| "loss": 0.3396, |
| "mean_token_accuracy": 0.887230110168457, |
| "num_tokens": 38296179.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.919668436050415, |
| "learning_rate": 3.2019586946045064e-05, |
| "loss": 0.3467, |
| "mean_token_accuracy": 0.8855098962783814, |
| "num_tokens": 38891168.0, |
| "step": 325 |
| }, |
| { |
| "epoch": 4.230769230769231, |
| "grad_norm": 0.9265286326408386, |
| "learning_rate": 3.11878707003765e-05, |
| "loss": 0.3185, |
| "mean_token_accuracy": 0.8956848978996277, |
| "num_tokens": 39468899.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 4.294871794871795, |
| "grad_norm": 1.013704538345337, |
| "learning_rate": 3.0375611129499047e-05, |
| "loss": 0.3393, |
| "mean_token_accuracy": 0.887428879737854, |
| "num_tokens": 40045289.0, |
| "step": 335 |
| }, |
| { |
| "epoch": 4.358974358974359, |
| "grad_norm": 0.8942686319351196, |
| "learning_rate": 2.9583815675246995e-05, |
| "loss": 0.3442, |
| "mean_token_accuracy": 0.8858568906784058, |
| "num_tokens": 40657913.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 4.423076923076923, |
| "grad_norm": 0.809492826461792, |
| "learning_rate": 2.881346639790431e-05, |
| "loss": 0.3477, |
| "mean_token_accuracy": 0.8856797218322754, |
| "num_tokens": 41294606.0, |
| "step": 345 |
| }, |
| { |
| "epoch": 4.487179487179487, |
| "grad_norm": 1.0839533805847168, |
| "learning_rate": 2.806551875815971e-05, |
| "loss": 0.3279, |
| "mean_token_accuracy": 0.8904742956161499, |
| "num_tokens": 41867505.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 4.551282051282051, |
| "grad_norm": 0.954156756401062, |
| "learning_rate": 2.7340900432053174e-05, |
| "loss": 0.3254, |
| "mean_token_accuracy": 0.8918370008468628, |
| "num_tokens": 42442620.0, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.615384615384615, |
| "grad_norm": 0.9666773676872253, |
| "learning_rate": 2.6640510160383592e-05, |
| "loss": 0.3512, |
| "mean_token_accuracy": 0.8855862617492676, |
| "num_tokens": 43046927.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.67948717948718, |
| "grad_norm": 0.95964115858078, |
| "learning_rate": 2.5965216634004592e-05, |
| "loss": 0.3205, |
| "mean_token_accuracy": 0.894673228263855, |
| "num_tokens": 43647447.0, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.743589743589744, |
| "grad_norm": 0.83061683177948, |
| "learning_rate": 2.531585741639134e-05, |
| "loss": 0.34, |
| "mean_token_accuracy": 0.8864041805267334, |
| "num_tokens": 44248341.0, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.8076923076923075, |
| "grad_norm": 0.938734769821167, |
| "learning_rate": 2.4693237904814327e-05, |
| "loss": 0.3431, |
| "mean_token_accuracy": 0.8860084414482117, |
| "num_tokens": 44831212.0, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.871794871794872, |
| "grad_norm": 0.8429780602455139, |
| "learning_rate": 2.409813033140881e-05, |
| "loss": 0.3622, |
| "mean_token_accuracy": 0.8836195230484009, |
| "num_tokens": 45451241.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.935897435897436, |
| "grad_norm": 0.7964617013931274, |
| "learning_rate": 2.353127280537887e-05, |
| "loss": 0.3308, |
| "mean_token_accuracy": 0.8908187031745911, |
| "num_tokens": 46081967.0, |
| "step": 385 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.9384036660194397, |
| "learning_rate": 2.2993368397523894e-05, |
| "loss": 0.3185, |
| "mean_token_accuracy": 0.8960811138153076, |
| "num_tokens": 46678966.0, |
| "step": 390 |
| }, |
| { |
| "epoch": 5.064102564102564, |
| "grad_norm": 0.8545572757720947, |
| "learning_rate": 2.2485084268223112e-05, |
| "loss": 0.3062, |
| "mean_token_accuracy": 0.896678626537323, |
| "num_tokens": 47288311.0, |
| "step": 395 |
| }, |
| { |
| "epoch": 5.128205128205128, |
| "grad_norm": 1.012990117073059, |
| "learning_rate": 2.200705083995957e-05, |
| "loss": 0.2563, |
| "mean_token_accuracy": 0.9135300159454346, |
| "num_tokens": 47864783.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 5.1923076923076925, |
| "grad_norm": 0.9687543511390686, |
| "learning_rate": 2.1559861015410074e-05, |
| "loss": 0.2807, |
| "mean_token_accuracy": 0.9048009753227234, |
| "num_tokens": 48487766.0, |
| "step": 405 |
| }, |
| { |
| "epoch": 5.256410256410256, |
| "grad_norm": 1.2796342372894287, |
| "learning_rate": 2.114406944207067e-05, |
| "loss": 0.2496, |
| "mean_token_accuracy": 0.9177995443344116, |
| "num_tokens": 49060095.0, |
| "step": 410 |
| }, |
| { |
| "epoch": 5.32051282051282, |
| "grad_norm": 1.0527794361114502, |
| "learning_rate": 2.076019182432989e-05, |
| "loss": 0.2988, |
| "mean_token_accuracy": 0.8987082719802857, |
| "num_tokens": 49645480.0, |
| "step": 415 |
| }, |
| { |
| "epoch": 5.384615384615385, |
| "grad_norm": 1.281862497329712, |
| "learning_rate": 2.040870428384297e-05, |
| "loss": 0.2788, |
| "mean_token_accuracy": 0.9077161908149719, |
| "num_tokens": 50275187.0, |
| "step": 420 |
| }, |
| { |
| "epoch": 5.448717948717949, |
| "grad_norm": 1.0032382011413574, |
| "learning_rate": 2.0090042769000326e-05, |
| "loss": 0.2803, |
| "mean_token_accuracy": 0.9113660097122193, |
| "num_tokens": 50835404.0, |
| "step": 425 |
| }, |
| { |
| "epoch": 5.512820512820513, |
| "grad_norm": 1.255702018737793, |
| "learning_rate": 1.980460251422275e-05, |
| "loss": 0.2967, |
| "mean_token_accuracy": 0.9015124201774597, |
| "num_tokens": 51462902.0, |
| "step": 430 |
| }, |
| { |
| "epoch": 5.576923076923077, |
| "grad_norm": 0.9799900054931641, |
| "learning_rate": 1.9552737549753983e-05, |
| "loss": 0.2998, |
| "mean_token_accuracy": 0.901181423664093, |
| "num_tokens": 52069408.0, |
| "step": 435 |
| }, |
| { |
| "epoch": 5.641025641025641, |
| "grad_norm": 1.036237120628357, |
| "learning_rate": 1.93347602625586e-05, |
| "loss": 0.2895, |
| "mean_token_accuracy": 0.9067635297775268, |
| "num_tokens": 52677193.0, |
| "step": 440 |
| }, |
| { |
| "epoch": 5.705128205128205, |
| "grad_norm": 0.9285158514976501, |
| "learning_rate": 1.9150941008869924e-05, |
| "loss": 0.295, |
| "mean_token_accuracy": 0.9037184834480285, |
| "num_tokens": 53250395.0, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.769230769230769, |
| "grad_norm": 0.99971604347229, |
| "learning_rate": 1.9001507778868364e-05, |
| "loss": 0.2844, |
| "mean_token_accuracy": 0.9065512895584107, |
| "num_tokens": 53810139.0, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.9049599766731262, |
| "learning_rate": 1.8886645913906293e-05, |
| "loss": 0.2827, |
| "mean_token_accuracy": 0.9079169154167175, |
| "num_tokens": 54409863.0, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.897435897435898, |
| "grad_norm": 1.0282909870147705, |
| "learning_rate": 1.8806497876629978e-05, |
| "loss": 0.2898, |
| "mean_token_accuracy": 0.9056537866592407, |
| "num_tokens": 55008893.0, |
| "step": 460 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 468, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 6, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 9.401212234241147e+17, |
| "train_batch_size": 48, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
