| { | |
| "best_metric": 0.4301721155643463, | |
| "best_model_checkpoint": "./vit-skin-demo-v1/checkpoint-1200", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1284, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019859813084112152, | |
| "loss": 1.3761, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019704049844236762, | |
| "loss": 0.9425, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019548286604361372, | |
| "loss": 0.9102, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019392523364485982, | |
| "loss": 0.8726, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019236760124610595, | |
| "loss": 0.8659, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00019080996884735202, | |
| "loss": 0.8507, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00018925233644859814, | |
| "loss": 0.8446, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00018769470404984424, | |
| "loss": 0.941, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00018613707165109034, | |
| "loss": 0.7384, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00018457943925233644, | |
| "loss": 0.7377, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.7553058676654182, | |
| "eval_loss": 0.7304826974868774, | |
| "eval_runtime": 19.4854, | |
| "eval_samples_per_second": 82.215, | |
| "eval_steps_per_second": 10.315, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00018302180685358257, | |
| "loss": 0.7471, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00018146417445482867, | |
| "loss": 0.6848, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00017990654205607477, | |
| "loss": 0.7872, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00017834890965732087, | |
| "loss": 0.7769, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.000176791277258567, | |
| "loss": 0.7049, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00017523364485981307, | |
| "loss": 0.4937, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0001736760124610592, | |
| "loss": 0.7708, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0001721183800623053, | |
| "loss": 0.891, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00017056074766355142, | |
| "loss": 0.7497, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001690031152647975, | |
| "loss": 0.8988, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_accuracy": 0.7540574282147315, | |
| "eval_loss": 0.6799274682998657, | |
| "eval_runtime": 19.407, | |
| "eval_samples_per_second": 82.548, | |
| "eval_steps_per_second": 10.357, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016744548286604362, | |
| "loss": 0.6315, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00016588785046728972, | |
| "loss": 0.8089, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00016433021806853582, | |
| "loss": 0.6893, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00016277258566978192, | |
| "loss": 0.7498, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00016121495327102805, | |
| "loss": 0.8541, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00015965732087227415, | |
| "loss": 0.6608, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00015809968847352025, | |
| "loss": 0.6824, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00015654205607476635, | |
| "loss": 0.7043, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00015498442367601248, | |
| "loss": 0.6091, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.00015342679127725855, | |
| "loss": 0.7157, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_accuracy": 0.7771535580524345, | |
| "eval_loss": 0.6038566827774048, | |
| "eval_runtime": 19.6645, | |
| "eval_samples_per_second": 81.467, | |
| "eval_steps_per_second": 10.221, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00015186915887850468, | |
| "loss": 0.517, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00015031152647975078, | |
| "loss": 0.6309, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.0001487538940809969, | |
| "loss": 0.7726, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.000147196261682243, | |
| "loss": 0.6196, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0001456386292834891, | |
| "loss": 0.4591, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00014408099688473523, | |
| "loss": 0.5935, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0001425233644859813, | |
| "loss": 0.5606, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00014096573208722743, | |
| "loss": 0.5984, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00013940809968847353, | |
| "loss": 0.5357, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00013785046728971963, | |
| "loss": 0.5569, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.7578027465667915, | |
| "eval_loss": 0.6506312489509583, | |
| "eval_runtime": 18.5005, | |
| "eval_samples_per_second": 86.592, | |
| "eval_steps_per_second": 10.865, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013629283489096573, | |
| "loss": 0.5707, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00013473520249221186, | |
| "loss": 0.5719, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00013317757009345796, | |
| "loss": 0.5915, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.00013161993769470406, | |
| "loss": 0.5368, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00013006230529595016, | |
| "loss": 0.6008, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00012850467289719628, | |
| "loss": 0.5129, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00012694704049844238, | |
| "loss": 0.4853, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00012538940809968848, | |
| "loss": 0.6041, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00012383177570093458, | |
| "loss": 0.701, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0001222741433021807, | |
| "loss": 0.5342, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.7846441947565543, | |
| "eval_loss": 0.5928722620010376, | |
| "eval_runtime": 19.6731, | |
| "eval_samples_per_second": 81.431, | |
| "eval_steps_per_second": 10.217, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0001207165109034268, | |
| "loss": 0.6391, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00011915887850467291, | |
| "loss": 0.5824, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.00011760124610591901, | |
| "loss": 0.5496, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00011604361370716512, | |
| "loss": 0.709, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00011448598130841121, | |
| "loss": 0.6867, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00011292834890965734, | |
| "loss": 0.4701, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00011137071651090342, | |
| "loss": 0.6941, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00010981308411214955, | |
| "loss": 0.5546, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.00010825545171339564, | |
| "loss": 0.3854, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00010669781931464175, | |
| "loss": 0.6498, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.7952559300873908, | |
| "eval_loss": 0.5553368926048279, | |
| "eval_runtime": 18.8966, | |
| "eval_samples_per_second": 84.777, | |
| "eval_steps_per_second": 10.637, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00010514018691588785, | |
| "loss": 0.5885, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00010358255451713396, | |
| "loss": 0.5098, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00010202492211838006, | |
| "loss": 0.5097, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00010046728971962618, | |
| "loss": 0.5651, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.890965732087229e-05, | |
| "loss": 0.3913, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.735202492211839e-05, | |
| "loss": 0.5351, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.579439252336449e-05, | |
| "loss": 0.3227, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.42367601246106e-05, | |
| "loss": 0.4168, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.26791277258567e-05, | |
| "loss": 0.4675, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 9.112149532710282e-05, | |
| "loss": 0.4956, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_accuracy": 0.7921348314606742, | |
| "eval_loss": 0.5429254174232483, | |
| "eval_runtime": 18.9872, | |
| "eval_samples_per_second": 84.373, | |
| "eval_steps_per_second": 10.586, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.956386292834892e-05, | |
| "loss": 0.4092, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.800623052959502e-05, | |
| "loss": 0.4268, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 8.644859813084113e-05, | |
| "loss": 0.4388, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.489096573208723e-05, | |
| "loss": 0.3978, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.5106, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 8.177570093457944e-05, | |
| "loss": 0.4768, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8.021806853582556e-05, | |
| "loss": 0.5389, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.866043613707166e-05, | |
| "loss": 0.4351, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 7.710280373831776e-05, | |
| "loss": 0.4425, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 7.554517133956387e-05, | |
| "loss": 0.5216, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_accuracy": 0.8302122347066168, | |
| "eval_loss": 0.4704440236091614, | |
| "eval_runtime": 18.9227, | |
| "eval_samples_per_second": 84.66, | |
| "eval_steps_per_second": 10.622, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 7.398753894080997e-05, | |
| "loss": 0.4263, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 7.242990654205608e-05, | |
| "loss": 0.346, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 7.087227414330218e-05, | |
| "loss": 0.4664, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 6.93146417445483e-05, | |
| "loss": 0.3724, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.77570093457944e-05, | |
| "loss": 0.4354, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 6.61993769470405e-05, | |
| "loss": 0.534, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 6.464174454828661e-05, | |
| "loss": 0.5173, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 6.308411214953271e-05, | |
| "loss": 0.4208, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 6.152647975077882e-05, | |
| "loss": 0.4199, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 5.9968847352024924e-05, | |
| "loss": 0.3468, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_accuracy": 0.83270911360799, | |
| "eval_loss": 0.46694162487983704, | |
| "eval_runtime": 19.215, | |
| "eval_samples_per_second": 83.372, | |
| "eval_steps_per_second": 10.461, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.841121495327103e-05, | |
| "loss": 0.3722, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 5.685358255451714e-05, | |
| "loss": 0.3283, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 5.5295950155763244e-05, | |
| "loss": 0.4791, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 5.373831775700935e-05, | |
| "loss": 0.3859, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 5.218068535825546e-05, | |
| "loss": 0.2583, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 5.062305295950156e-05, | |
| "loss": 0.318, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 4.9065420560747664e-05, | |
| "loss": 0.3943, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.750778816199377e-05, | |
| "loss": 0.2449, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.595015576323988e-05, | |
| "loss": 0.204, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.4392523364485984e-05, | |
| "loss": 0.4862, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.8420724094881398, | |
| "eval_loss": 0.4615306556224823, | |
| "eval_runtime": 19.2799, | |
| "eval_samples_per_second": 83.092, | |
| "eval_steps_per_second": 10.425, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 4.283489096573209e-05, | |
| "loss": 0.2649, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.12772585669782e-05, | |
| "loss": 0.3136, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.97196261682243e-05, | |
| "loss": 0.2549, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 3.8161993769470404e-05, | |
| "loss": 0.2944, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.660436137071651e-05, | |
| "loss": 0.3809, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 3.504672897196262e-05, | |
| "loss": 0.2504, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.3489096573208724e-05, | |
| "loss": 0.4471, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.193146417445483e-05, | |
| "loss": 0.2706, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3.0373831775700934e-05, | |
| "loss": 0.2495, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.881619937694704e-05, | |
| "loss": 0.4018, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "eval_accuracy": 0.8458177278401997, | |
| "eval_loss": 0.4525544345378876, | |
| "eval_runtime": 18.7511, | |
| "eval_samples_per_second": 85.435, | |
| "eval_steps_per_second": 10.719, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 2.7258566978193147e-05, | |
| "loss": 0.2398, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 2.570093457943925e-05, | |
| "loss": 0.3493, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 2.414330218068536e-05, | |
| "loss": 0.1811, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 2.2585669781931463e-05, | |
| "loss": 0.2008, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 2.102803738317757e-05, | |
| "loss": 0.2805, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.9470404984423677e-05, | |
| "loss": 0.2142, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.7912772585669783e-05, | |
| "loss": 0.2079, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.6355140186915887e-05, | |
| "loss": 0.3097, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 1.4797507788161993e-05, | |
| "loss": 0.2846, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.3239875389408098e-05, | |
| "loss": 0.302, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_accuracy": 0.8558052434456929, | |
| "eval_loss": 0.4301721155643463, | |
| "eval_runtime": 19.5305, | |
| "eval_samples_per_second": 82.026, | |
| "eval_steps_per_second": 10.292, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 1.1682242990654207e-05, | |
| "loss": 0.3489, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.0124610591900312e-05, | |
| "loss": 0.3015, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.566978193146417e-06, | |
| "loss": 0.231, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 7.009345794392523e-06, | |
| "loss": 0.192, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 5.451713395638629e-06, | |
| "loss": 0.2904, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.894080996884735e-06, | |
| "loss": 0.412, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 2.3364485981308413e-06, | |
| "loss": 0.2304, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 7.788161993769471e-07, | |
| "loss": 0.293, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1284, | |
| "total_flos": 1.5895869226063626e+18, | |
| "train_loss": 0.5172773925036285, | |
| "train_runtime": 727.4962, | |
| "train_samples_per_second": 28.195, | |
| "train_steps_per_second": 1.765 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1284, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 1.5895869226063626e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |