| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 7400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.2912011474885726, | |
| "eval_loss": 5.149496078491211, | |
| "eval_runtime": 19.7373, | |
| "eval_samples_per_second": 59.228, | |
| "eval_steps_per_second": 1.875, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.33825854783938614, | |
| "eval_loss": 4.6761064529418945, | |
| "eval_runtime": 19.7373, | |
| "eval_samples_per_second": 59.228, | |
| "eval_steps_per_second": 1.875, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.37122292212112573, | |
| "eval_loss": 4.382660388946533, | |
| "eval_runtime": 19.7317, | |
| "eval_samples_per_second": 59.245, | |
| "eval_steps_per_second": 1.875, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.662162162162162e-05, | |
| "loss": 4.9816, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3959613241050367, | |
| "eval_loss": 4.176214694976807, | |
| "eval_runtime": 19.7386, | |
| "eval_samples_per_second": 59.224, | |
| "eval_steps_per_second": 1.874, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.41364025196360527, | |
| "eval_loss": 4.029497146606445, | |
| "eval_runtime": 19.7378, | |
| "eval_samples_per_second": 59.226, | |
| "eval_steps_per_second": 1.875, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4276382300334396, | |
| "eval_loss": 3.915071725845337, | |
| "eval_runtime": 19.7391, | |
| "eval_samples_per_second": 59.223, | |
| "eval_steps_per_second": 1.874, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 4.324324324324325e-05, | |
| "loss": 3.9068, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.43997718848018247, | |
| "eval_loss": 3.829800844192505, | |
| "eval_runtime": 19.73, | |
| "eval_samples_per_second": 59.25, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1036 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.450026354217971, | |
| "eval_loss": 3.7551472187042236, | |
| "eval_runtime": 19.7353, | |
| "eval_samples_per_second": 59.234, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1184 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.45856339269512925, | |
| "eval_loss": 3.6967544555664062, | |
| "eval_runtime": 19.735, | |
| "eval_samples_per_second": 59.235, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4668152871745686, | |
| "eval_loss": 3.6389007568359375, | |
| "eval_runtime": 19.7369, | |
| "eval_samples_per_second": 59.229, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 3.986486486486487e-05, | |
| "loss": 3.3777, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4742895162056839, | |
| "eval_loss": 3.5959553718566895, | |
| "eval_runtime": 19.7345, | |
| "eval_samples_per_second": 59.236, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1628 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4803034623393905, | |
| "eval_loss": 3.5559630393981934, | |
| "eval_runtime": 19.7359, | |
| "eval_samples_per_second": 59.232, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1776 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.48679264846929515, | |
| "eval_loss": 3.537980794906616, | |
| "eval_runtime": 19.7353, | |
| "eval_samples_per_second": 59.234, | |
| "eval_steps_per_second": 1.875, | |
| "step": 1924 | |
| }, | |
| { | |
| "epoch": 13.51, | |
| "learning_rate": 3.648648648648649e-05, | |
| "loss": 3.0075, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.49136359316000033, | |
| "eval_loss": 3.484649181365967, | |
| "eval_runtime": 20.0887, | |
| "eval_samples_per_second": 58.192, | |
| "eval_steps_per_second": 1.842, | |
| "step": 2072 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.49689365857030526, | |
| "eval_loss": 3.4657890796661377, | |
| "eval_runtime": 19.7322, | |
| "eval_samples_per_second": 59.243, | |
| "eval_steps_per_second": 1.875, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.501041207628034, | |
| "eval_loss": 3.4555246829986572, | |
| "eval_runtime": 19.7332, | |
| "eval_samples_per_second": 59.24, | |
| "eval_steps_per_second": 1.875, | |
| "step": 2368 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 3.310810810810811e-05, | |
| "loss": 2.7329, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5052838046850023, | |
| "eval_loss": 3.4299707412719727, | |
| "eval_runtime": 19.7326, | |
| "eval_samples_per_second": 59.242, | |
| "eval_steps_per_second": 1.875, | |
| "step": 2516 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5079710708453223, | |
| "eval_loss": 3.420835494995117, | |
| "eval_runtime": 19.737, | |
| "eval_samples_per_second": 59.229, | |
| "eval_steps_per_second": 1.875, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.5120667755398294, | |
| "eval_loss": 3.425023078918457, | |
| "eval_runtime": 19.7397, | |
| "eval_samples_per_second": 59.221, | |
| "eval_steps_per_second": 1.874, | |
| "step": 2812 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5146849158825206, | |
| "eval_loss": 3.3964016437530518, | |
| "eval_runtime": 19.7401, | |
| "eval_samples_per_second": 59.22, | |
| "eval_steps_per_second": 1.874, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 2.9729729729729733e-05, | |
| "loss": 2.5153, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.5181498474911649, | |
| "eval_loss": 3.3892605304718018, | |
| "eval_runtime": 19.7415, | |
| "eval_samples_per_second": 59.215, | |
| "eval_steps_per_second": 1.874, | |
| "step": 3108 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.5204396402001192, | |
| "eval_loss": 3.3914339542388916, | |
| "eval_runtime": 19.7345, | |
| "eval_samples_per_second": 59.236, | |
| "eval_steps_per_second": 1.875, | |
| "step": 3256 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.5228590438171277, | |
| "eval_loss": 3.3818881511688232, | |
| "eval_runtime": 19.7369, | |
| "eval_samples_per_second": 59.229, | |
| "eval_steps_per_second": 1.875, | |
| "step": 3404 | |
| }, | |
| { | |
| "epoch": 23.65, | |
| "learning_rate": 2.635135135135135e-05, | |
| "loss": 2.336, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5246649558026804, | |
| "eval_loss": 3.3786017894744873, | |
| "eval_runtime": 19.7359, | |
| "eval_samples_per_second": 59.232, | |
| "eval_steps_per_second": 1.875, | |
| "step": 3552 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.5267128081499339, | |
| "eval_loss": 3.374890089035034, | |
| "eval_runtime": 19.7329, | |
| "eval_samples_per_second": 59.241, | |
| "eval_steps_per_second": 1.875, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.5287001754067623, | |
| "eval_loss": 3.377443552017212, | |
| "eval_runtime": 19.7328, | |
| "eval_samples_per_second": 59.241, | |
| "eval_steps_per_second": 1.875, | |
| "step": 3848 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.5302555063034105, | |
| "eval_loss": 3.3700437545776367, | |
| "eval_runtime": 19.7375, | |
| "eval_samples_per_second": 59.227, | |
| "eval_steps_per_second": 1.875, | |
| "step": 3996 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "learning_rate": 2.2972972972972976e-05, | |
| "loss": 2.1918, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5320614182889633, | |
| "eval_loss": 3.372175693511963, | |
| "eval_runtime": 19.7333, | |
| "eval_samples_per_second": 59.24, | |
| "eval_steps_per_second": 1.875, | |
| "step": 4144 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.5339623782737555, | |
| "eval_loss": 3.3728740215301514, | |
| "eval_runtime": 19.7294, | |
| "eval_samples_per_second": 59.252, | |
| "eval_steps_per_second": 1.875, | |
| "step": 4292 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.5349819840837805, | |
| "eval_loss": 3.3896090984344482, | |
| "eval_runtime": 19.7317, | |
| "eval_samples_per_second": 59.245, | |
| "eval_steps_per_second": 1.875, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "learning_rate": 1.9594594594594595e-05, | |
| "loss": 2.0717, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.5367446924333152, | |
| "eval_loss": 3.37764048576355, | |
| "eval_runtime": 19.7293, | |
| "eval_samples_per_second": 59.252, | |
| "eval_steps_per_second": 1.875, | |
| "step": 4588 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5384901193284427, | |
| "eval_loss": 3.384190797805786, | |
| "eval_runtime": 19.7297, | |
| "eval_samples_per_second": 59.251, | |
| "eval_steps_per_second": 1.875, | |
| "step": 4736 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.5399244800442405, | |
| "eval_loss": 3.38201642036438, | |
| "eval_runtime": 19.7316, | |
| "eval_samples_per_second": 59.245, | |
| "eval_steps_per_second": 1.875, | |
| "step": 4884 | |
| }, | |
| { | |
| "epoch": 33.78, | |
| "learning_rate": 1.6216216216216218e-05, | |
| "loss": 1.9814, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.540391079313235, | |
| "eval_loss": 3.393296241760254, | |
| "eval_runtime": 19.7393, | |
| "eval_samples_per_second": 59.222, | |
| "eval_steps_per_second": 1.874, | |
| "step": 5032 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.5410564153079123, | |
| "eval_loss": 3.386063814163208, | |
| "eval_runtime": 19.7351, | |
| "eval_samples_per_second": 59.234, | |
| "eval_steps_per_second": 1.875, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5425253389325245, | |
| "eval_loss": 3.3877809047698975, | |
| "eval_runtime": 19.7289, | |
| "eval_samples_per_second": 59.253, | |
| "eval_steps_per_second": 1.875, | |
| "step": 5328 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.5431129083823695, | |
| "eval_loss": 3.3903427124023438, | |
| "eval_runtime": 19.7301, | |
| "eval_samples_per_second": 59.25, | |
| "eval_steps_per_second": 1.875, | |
| "step": 5476 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 1.2837837837837838e-05, | |
| "loss": 1.9049, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.5439856218299332, | |
| "eval_loss": 3.3847618103027344, | |
| "eval_runtime": 19.731, | |
| "eval_samples_per_second": 59.247, | |
| "eval_steps_per_second": 1.875, | |
| "step": 5624 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.5446855207334249, | |
| "eval_loss": 3.396543502807617, | |
| "eval_runtime": 19.7337, | |
| "eval_samples_per_second": 59.239, | |
| "eval_steps_per_second": 1.875, | |
| "step": 5772 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.5453249345464914, | |
| "eval_loss": 3.403301477432251, | |
| "eval_runtime": 19.7292, | |
| "eval_samples_per_second": 59.252, | |
| "eval_steps_per_second": 1.875, | |
| "step": 5920 | |
| }, | |
| { | |
| "epoch": 40.54, | |
| "learning_rate": 9.45945945945946e-06, | |
| "loss": 1.8441, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.5456187192714139, | |
| "eval_loss": 3.4074459075927734, | |
| "eval_runtime": 19.7354, | |
| "eval_samples_per_second": 59.234, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6068 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.5462322109028696, | |
| "eval_loss": 3.4045963287353516, | |
| "eval_runtime": 19.7291, | |
| "eval_samples_per_second": 59.253, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6216 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.5467333730806785, | |
| "eval_loss": 3.4119956493377686, | |
| "eval_runtime": 19.7304, | |
| "eval_samples_per_second": 59.249, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6364 | |
| }, | |
| { | |
| "epoch": 43.92, | |
| "learning_rate": 6.081081081081082e-06, | |
| "loss": 1.804, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.5466901694446604, | |
| "eval_loss": 3.4043545722961426, | |
| "eval_runtime": 19.7301, | |
| "eval_samples_per_second": 59.25, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6512 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.5471913316224694, | |
| "eval_loss": 3.4124903678894043, | |
| "eval_runtime": 19.7367, | |
| "eval_samples_per_second": 59.23, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6660 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.5476665716186674, | |
| "eval_loss": 3.4115209579467773, | |
| "eval_runtime": 19.7299, | |
| "eval_samples_per_second": 59.25, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6808 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.547675212345871, | |
| "eval_loss": 3.4069876670837402, | |
| "eval_runtime": 19.7339, | |
| "eval_samples_per_second": 59.238, | |
| "eval_steps_per_second": 1.875, | |
| "step": 6956 | |
| }, | |
| { | |
| "epoch": 47.3, | |
| "learning_rate": 2.702702702702703e-06, | |
| "loss": 1.7744, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.5477875417995178, | |
| "eval_loss": 3.420271396636963, | |
| "eval_runtime": 19.7291, | |
| "eval_samples_per_second": 59.253, | |
| "eval_steps_per_second": 1.875, | |
| "step": 7104 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.5479430748891827, | |
| "eval_loss": 3.4173614978790283, | |
| "eval_runtime": 19.7304, | |
| "eval_samples_per_second": 59.249, | |
| "eval_steps_per_second": 1.875, | |
| "step": 7252 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.5480813265244403, | |
| "eval_loss": 3.4181294441223145, | |
| "eval_runtime": 19.7276, | |
| "eval_samples_per_second": 59.257, | |
| "eval_steps_per_second": 1.876, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 7400, | |
| "total_flos": 1.00265577216e+17, | |
| "train_loss": 2.556491814690667, | |
| "train_runtime": 9037.2432, | |
| "train_samples_per_second": 26.103, | |
| "train_steps_per_second": 0.819 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.00265577216e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |