{ "best_metric": null, "best_model_checkpoint": null, "epoch": 10.0, "eval_steps": 1755, "global_step": 3510, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.07, "learning_rate": 1.985754985754986e-05, "loss": 0.5449, "step": 25 }, { "epoch": 0.14, "learning_rate": 1.9715099715099716e-05, "loss": 0.3107, "step": 50 }, { "epoch": 0.21, "learning_rate": 1.9572649572649574e-05, "loss": 0.2798, "step": 75 }, { "epoch": 0.28, "learning_rate": 1.943019943019943e-05, "loss": 0.2448, "step": 100 }, { "epoch": 0.36, "learning_rate": 1.928774928774929e-05, "loss": 0.2447, "step": 125 }, { "epoch": 0.43, "learning_rate": 1.914529914529915e-05, "loss": 0.2314, "step": 150 }, { "epoch": 0.5, "learning_rate": 1.9002849002849003e-05, "loss": 0.1924, "step": 175 }, { "epoch": 0.57, "learning_rate": 1.8860398860398864e-05, "loss": 0.2063, "step": 200 }, { "epoch": 0.64, "learning_rate": 1.8717948717948718e-05, "loss": 0.1999, "step": 225 }, { "epoch": 0.71, "learning_rate": 1.8575498575498575e-05, "loss": 0.1794, "step": 250 }, { "epoch": 0.78, "learning_rate": 1.8433048433048436e-05, "loss": 0.1884, "step": 275 }, { "epoch": 0.85, "learning_rate": 1.829059829059829e-05, "loss": 0.1819, "step": 300 }, { "epoch": 0.93, "learning_rate": 1.814814814814815e-05, "loss": 0.1682, "step": 325 }, { "epoch": 1.0, "learning_rate": 1.8005698005698008e-05, "loss": 0.1808, "step": 350 }, { "epoch": 1.07, "learning_rate": 1.7863247863247866e-05, "loss": 0.1332, "step": 375 }, { "epoch": 1.14, "learning_rate": 1.7720797720797723e-05, "loss": 0.1152, "step": 400 }, { "epoch": 1.21, "learning_rate": 1.757834757834758e-05, "loss": 0.1308, "step": 425 }, { "epoch": 1.28, "learning_rate": 1.7435897435897438e-05, "loss": 0.1344, "step": 450 }, { "epoch": 1.35, "learning_rate": 1.7293447293447295e-05, "loss": 0.1267, "step": 475 }, { "epoch": 1.42, "learning_rate": 1.7150997150997152e-05, "loss": 0.1333, "step": 500 }, { "epoch": 1.5, "learning_rate": 1.700854700854701e-05, "loss": 0.1278, "step": 525 }, { "epoch": 1.57, "learning_rate": 1.6866096866096867e-05, "loss": 0.1326, "step": 550 }, { "epoch": 1.64, "learning_rate": 1.6723646723646725e-05, "loss": 0.1148, "step": 575 }, { "epoch": 1.71, "learning_rate": 1.6581196581196585e-05, "loss": 0.1207, "step": 600 }, { "epoch": 1.78, "learning_rate": 1.643874643874644e-05, "loss": 0.12, "step": 625 }, { "epoch": 1.85, "learning_rate": 1.6296296296296297e-05, "loss": 0.1129, "step": 650 }, { "epoch": 1.92, "learning_rate": 1.6153846153846154e-05, "loss": 0.1295, "step": 675 }, { "epoch": 1.99, "learning_rate": 1.601139601139601e-05, "loss": 0.1244, "step": 700 }, { "epoch": 2.07, "learning_rate": 1.5868945868945872e-05, "loss": 0.0947, "step": 725 }, { "epoch": 2.14, "learning_rate": 1.5726495726495726e-05, "loss": 0.0791, "step": 750 }, { "epoch": 2.21, "learning_rate": 1.5584045584045587e-05, "loss": 0.0846, "step": 775 }, { "epoch": 2.28, "learning_rate": 1.5441595441595444e-05, "loss": 0.083, "step": 800 }, { "epoch": 2.35, "learning_rate": 1.5299145299145298e-05, "loss": 0.0859, "step": 825 }, { "epoch": 2.42, "learning_rate": 1.5156695156695157e-05, "loss": 0.0851, "step": 850 }, { "epoch": 2.49, "learning_rate": 1.5014245014245015e-05, "loss": 0.0916, "step": 875 }, { "epoch": 2.56, "learning_rate": 1.4871794871794874e-05, "loss": 0.0822, "step": 900 }, { "epoch": 2.64, "learning_rate": 1.472934472934473e-05, "loss": 0.0933, "step": 925 }, { "epoch": 2.71, "learning_rate": 1.4586894586894588e-05, "loss": 0.0829, "step": 950 }, { "epoch": 2.78, "learning_rate": 1.4444444444444446e-05, "loss": 0.0908, "step": 975 }, { "epoch": 2.85, "learning_rate": 1.4301994301994305e-05, "loss": 0.0911, "step": 1000 }, { "epoch": 2.92, "learning_rate": 1.415954415954416e-05, "loss": 0.0851, "step": 1025 }, { "epoch": 2.99, "learning_rate": 1.4017094017094018e-05, "loss": 0.0958, "step": 1050 }, { "epoch": 3.06, "learning_rate": 1.3874643874643875e-05, "loss": 0.0588, "step": 1075 }, { "epoch": 3.13, "learning_rate": 1.3732193732193733e-05, "loss": 0.0566, "step": 1100 }, { "epoch": 3.21, "learning_rate": 1.3589743589743592e-05, "loss": 0.0573, "step": 1125 }, { "epoch": 3.28, "learning_rate": 1.3447293447293447e-05, "loss": 0.0603, "step": 1150 }, { "epoch": 3.35, "learning_rate": 1.3304843304843306e-05, "loss": 0.0753, "step": 1175 }, { "epoch": 3.42, "learning_rate": 1.3162393162393164e-05, "loss": 0.0577, "step": 1200 }, { "epoch": 3.49, "learning_rate": 1.301994301994302e-05, "loss": 0.0678, "step": 1225 }, { "epoch": 3.56, "learning_rate": 1.2877492877492879e-05, "loss": 0.0696, "step": 1250 }, { "epoch": 3.63, "learning_rate": 1.2735042735042736e-05, "loss": 0.067, "step": 1275 }, { "epoch": 3.7, "learning_rate": 1.2592592592592593e-05, "loss": 0.0658, "step": 1300 }, { "epoch": 3.77, "learning_rate": 1.245014245014245e-05, "loss": 0.0701, "step": 1325 }, { "epoch": 3.85, "learning_rate": 1.230769230769231e-05, "loss": 0.0634, "step": 1350 }, { "epoch": 3.92, "learning_rate": 1.2165242165242165e-05, "loss": 0.0711, "step": 1375 }, { "epoch": 3.99, "learning_rate": 1.2022792022792024e-05, "loss": 0.0667, "step": 1400 }, { "epoch": 4.06, "learning_rate": 1.1880341880341882e-05, "loss": 0.048, "step": 1425 }, { "epoch": 4.13, "learning_rate": 1.1737891737891738e-05, "loss": 0.0461, "step": 1450 }, { "epoch": 4.2, "learning_rate": 1.1595441595441597e-05, "loss": 0.0418, "step": 1475 }, { "epoch": 4.27, "learning_rate": 1.1452991452991454e-05, "loss": 0.0578, "step": 1500 }, { "epoch": 4.34, "learning_rate": 1.1310541310541311e-05, "loss": 0.0463, "step": 1525 }, { "epoch": 4.42, "learning_rate": 1.1168091168091169e-05, "loss": 0.0458, "step": 1550 }, { "epoch": 4.49, "learning_rate": 1.1025641025641028e-05, "loss": 0.0509, "step": 1575 }, { "epoch": 4.56, "learning_rate": 1.0883190883190883e-05, "loss": 0.0509, "step": 1600 }, { "epoch": 4.63, "learning_rate": 1.0740740740740742e-05, "loss": 0.0453, "step": 1625 }, { "epoch": 4.7, "learning_rate": 1.05982905982906e-05, "loss": 0.0425, "step": 1650 }, { "epoch": 4.77, "learning_rate": 1.0455840455840456e-05, "loss": 0.0596, "step": 1675 }, { "epoch": 4.84, "learning_rate": 1.0313390313390315e-05, "loss": 0.0461, "step": 1700 }, { "epoch": 4.91, "learning_rate": 1.0170940170940172e-05, "loss": 0.0561, "step": 1725 }, { "epoch": 4.99, "learning_rate": 1.002849002849003e-05, "loss": 0.0523, "step": 1750 }, { "epoch": 5.06, "learning_rate": 9.886039886039887e-06, "loss": 0.0352, "step": 1775 }, { "epoch": 5.13, "learning_rate": 9.743589743589744e-06, "loss": 0.036, "step": 1800 }, { "epoch": 5.2, "learning_rate": 9.601139601139601e-06, "loss": 0.036, "step": 1825 }, { "epoch": 5.27, "learning_rate": 9.458689458689459e-06, "loss": 0.0343, "step": 1850 }, { "epoch": 5.34, "learning_rate": 9.316239316239318e-06, "loss": 0.0332, "step": 1875 }, { "epoch": 5.41, "learning_rate": 9.173789173789175e-06, "loss": 0.0398, "step": 1900 }, { "epoch": 5.48, "learning_rate": 9.031339031339033e-06, "loss": 0.0412, "step": 1925 }, { "epoch": 5.56, "learning_rate": 8.888888888888888e-06, "loss": 0.0397, "step": 1950 }, { "epoch": 5.63, "learning_rate": 8.746438746438747e-06, "loss": 0.0421, "step": 1975 }, { "epoch": 5.7, "learning_rate": 8.603988603988605e-06, "loss": 0.0386, "step": 2000 }, { "epoch": 5.77, "learning_rate": 8.461538461538462e-06, "loss": 0.0405, "step": 2025 }, { "epoch": 5.84, "learning_rate": 8.31908831908832e-06, "loss": 0.0431, "step": 2050 }, { "epoch": 5.91, "learning_rate": 8.176638176638177e-06, "loss": 0.0464, "step": 2075 }, { "epoch": 5.98, "learning_rate": 8.034188034188036e-06, "loss": 0.0462, "step": 2100 }, { "epoch": 6.05, "learning_rate": 7.891737891737893e-06, "loss": 0.0261, "step": 2125 }, { "epoch": 6.13, "learning_rate": 7.749287749287749e-06, "loss": 0.0304, "step": 2150 }, { "epoch": 6.2, "learning_rate": 7.606837606837607e-06, "loss": 0.029, "step": 2175 }, { "epoch": 6.27, "learning_rate": 7.4643874643874645e-06, "loss": 0.0316, "step": 2200 }, { "epoch": 6.34, "learning_rate": 7.321937321937323e-06, "loss": 0.0306, "step": 2225 }, { "epoch": 6.41, "learning_rate": 7.17948717948718e-06, "loss": 0.035, "step": 2250 }, { "epoch": 6.48, "learning_rate": 7.0370370370370375e-06, "loss": 0.0245, "step": 2275 }, { "epoch": 6.55, "learning_rate": 6.894586894586896e-06, "loss": 0.0207, "step": 2300 }, { "epoch": 6.62, "learning_rate": 6.752136752136753e-06, "loss": 0.0314, "step": 2325 }, { "epoch": 6.7, "learning_rate": 6.60968660968661e-06, "loss": 0.0291, "step": 2350 }, { "epoch": 6.77, "learning_rate": 6.467236467236467e-06, "loss": 0.0331, "step": 2375 }, { "epoch": 6.84, "learning_rate": 6.324786324786325e-06, "loss": 0.0297, "step": 2400 }, { "epoch": 6.91, "learning_rate": 6.1823361823361825e-06, "loss": 0.0321, "step": 2425 }, { "epoch": 6.98, "learning_rate": 6.039886039886041e-06, "loss": 0.0313, "step": 2450 }, { "epoch": 7.05, "learning_rate": 5.897435897435898e-06, "loss": 0.0228, "step": 2475 }, { "epoch": 7.12, "learning_rate": 5.7549857549857555e-06, "loss": 0.0167, "step": 2500 }, { "epoch": 7.19, "learning_rate": 5.612535612535614e-06, "loss": 0.0193, "step": 2525 }, { "epoch": 7.26, "learning_rate": 5.470085470085471e-06, "loss": 0.0244, "step": 2550 }, { "epoch": 7.34, "learning_rate": 5.327635327635328e-06, "loss": 0.0234, "step": 2575 }, { "epoch": 7.41, "learning_rate": 5.185185185185185e-06, "loss": 0.0232, "step": 2600 }, { "epoch": 7.48, "learning_rate": 5.042735042735043e-06, "loss": 0.025, "step": 2625 }, { "epoch": 7.55, "learning_rate": 4.9002849002849006e-06, "loss": 0.0265, "step": 2650 }, { "epoch": 7.62, "learning_rate": 4.757834757834758e-06, "loss": 0.0251, "step": 2675 }, { "epoch": 7.69, "learning_rate": 4.615384615384616e-06, "loss": 0.0269, "step": 2700 }, { "epoch": 7.76, "learning_rate": 4.4729344729344735e-06, "loss": 0.0264, "step": 2725 }, { "epoch": 7.83, "learning_rate": 4.330484330484331e-06, "loss": 0.0211, "step": 2750 }, { "epoch": 7.91, "learning_rate": 4.188034188034188e-06, "loss": 0.03, "step": 2775 }, { "epoch": 7.98, "learning_rate": 4.0455840455840465e-06, "loss": 0.0224, "step": 2800 }, { "epoch": 8.05, "learning_rate": 3.903133903133903e-06, "loss": 0.0233, "step": 2825 }, { "epoch": 8.12, "learning_rate": 3.760683760683761e-06, "loss": 0.0213, "step": 2850 }, { "epoch": 8.19, "learning_rate": 3.6182336182336186e-06, "loss": 0.0157, "step": 2875 }, { "epoch": 8.26, "learning_rate": 3.4757834757834764e-06, "loss": 0.0171, "step": 2900 }, { "epoch": 8.33, "learning_rate": 3.3333333333333333e-06, "loss": 0.0189, "step": 2925 }, { "epoch": 8.4, "learning_rate": 3.190883190883191e-06, "loss": 0.0154, "step": 2950 }, { "epoch": 8.48, "learning_rate": 3.048433048433049e-06, "loss": 0.0178, "step": 2975 }, { "epoch": 8.55, "learning_rate": 2.9059829059829063e-06, "loss": 0.0257, "step": 3000 }, { "epoch": 8.62, "learning_rate": 2.7635327635327636e-06, "loss": 0.0166, "step": 3025 }, { "epoch": 8.69, "learning_rate": 2.6210826210826214e-06, "loss": 0.0202, "step": 3050 }, { "epoch": 8.76, "learning_rate": 2.478632478632479e-06, "loss": 0.0176, "step": 3075 }, { "epoch": 8.83, "learning_rate": 2.336182336182336e-06, "loss": 0.0128, "step": 3100 }, { "epoch": 8.9, "learning_rate": 2.193732193732194e-06, "loss": 0.0263, "step": 3125 }, { "epoch": 8.97, "learning_rate": 2.0512820512820513e-06, "loss": 0.025, "step": 3150 }, { "epoch": 9.05, "learning_rate": 1.908831908831909e-06, "loss": 0.0151, "step": 3175 }, { "epoch": 9.12, "learning_rate": 1.7663817663817665e-06, "loss": 0.015, "step": 3200 }, { "epoch": 9.19, "learning_rate": 1.623931623931624e-06, "loss": 0.0125, "step": 3225 }, { "epoch": 9.26, "learning_rate": 1.4814814814814815e-06, "loss": 0.0116, "step": 3250 }, { "epoch": 9.33, "learning_rate": 1.3390313390313392e-06, "loss": 0.0197, "step": 3275 }, { "epoch": 9.4, "learning_rate": 1.1965811965811968e-06, "loss": 0.0196, "step": 3300 }, { "epoch": 9.47, "learning_rate": 1.0541310541310542e-06, "loss": 0.0183, "step": 3325 }, { "epoch": 9.54, "learning_rate": 9.116809116809118e-07, "loss": 0.0171, "step": 3350 }, { "epoch": 9.62, "learning_rate": 7.692307692307694e-07, "loss": 0.0139, "step": 3375 }, { "epoch": 9.69, "learning_rate": 6.267806267806268e-07, "loss": 0.0158, "step": 3400 }, { "epoch": 9.76, "learning_rate": 4.843304843304843e-07, "loss": 0.0178, "step": 3425 }, { "epoch": 9.83, "learning_rate": 3.4188034188034194e-07, "loss": 0.0112, "step": 3450 }, { "epoch": 9.9, "learning_rate": 1.9943019943019944e-07, "loss": 0.0146, "step": 3475 }, { "epoch": 9.97, "learning_rate": 5.6980056980056986e-08, "loss": 0.0108, "step": 3500 }, { "epoch": 10.0, "step": 3510, "total_flos": 4.43006661686016e+16, "train_loss": 0.0692601463139227, "train_runtime": 1373.3603, "train_samples_per_second": 490.396, "train_steps_per_second": 2.556 } ], "logging_steps": 25, "max_steps": 3510, "num_train_epochs": 10, "save_steps": 500, "total_flos": 4.43006661686016e+16, "trial_name": null, "trial_params": null }