| { | |
| "best_metric": 0.22059296071529388, | |
| "best_model_checkpoint": "finetuned-indian-food/checkpoint-1300", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1332, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03003003003003003, | |
| "grad_norm": 1.8055429458618164, | |
| "learning_rate": 0.0001984984984984985, | |
| "loss": 2.8033, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06006006006006006, | |
| "grad_norm": 1.9660130739212036, | |
| "learning_rate": 0.00019699699699699701, | |
| "loss": 2.4877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09009009009009009, | |
| "grad_norm": 2.4530346393585205, | |
| "learning_rate": 0.0001954954954954955, | |
| "loss": 2.1957, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12012012012012012, | |
| "grad_norm": 2.1654393672943115, | |
| "learning_rate": 0.00019399399399399402, | |
| "loss": 1.8884, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15015015015015015, | |
| "grad_norm": 1.9584170579910278, | |
| "learning_rate": 0.0001924924924924925, | |
| "loss": 1.7159, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18018018018018017, | |
| "grad_norm": 2.834153175354004, | |
| "learning_rate": 0.000190990990990991, | |
| "loss": 1.4741, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21021021021021022, | |
| "grad_norm": 3.5546669960021973, | |
| "learning_rate": 0.0001894894894894895, | |
| "loss": 1.2572, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24024024024024024, | |
| "grad_norm": 2.346996545791626, | |
| "learning_rate": 0.000187987987987988, | |
| "loss": 1.1259, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 2.314241886138916, | |
| "learning_rate": 0.0001864864864864865, | |
| "loss": 1.118, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3003003003003003, | |
| "grad_norm": 3.745007038116455, | |
| "learning_rate": 0.000184984984984985, | |
| "loss": 1.1092, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3003003003003003, | |
| "eval_accuracy": 0.822529224229543, | |
| "eval_loss": 1.0006803274154663, | |
| "eval_runtime": 29.7003, | |
| "eval_samples_per_second": 31.683, | |
| "eval_steps_per_second": 3.973, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3303303303303303, | |
| "grad_norm": 3.29766845703125, | |
| "learning_rate": 0.0001834834834834835, | |
| "loss": 0.9543, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.36036036036036034, | |
| "grad_norm": 1.7774616479873657, | |
| "learning_rate": 0.000181981981981982, | |
| "loss": 0.9778, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.39039039039039036, | |
| "grad_norm": 4.704737186431885, | |
| "learning_rate": 0.0001804804804804805, | |
| "loss": 0.9743, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.42042042042042044, | |
| "grad_norm": 3.2633509635925293, | |
| "learning_rate": 0.00017897897897897898, | |
| "loss": 0.8942, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.45045045045045046, | |
| "grad_norm": 3.4661707878112793, | |
| "learning_rate": 0.0001774774774774775, | |
| "loss": 0.711, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4804804804804805, | |
| "grad_norm": 3.7788476943969727, | |
| "learning_rate": 0.000175975975975976, | |
| "loss": 0.8988, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5105105105105106, | |
| "grad_norm": 3.419523000717163, | |
| "learning_rate": 0.0001744744744744745, | |
| "loss": 0.7006, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 2.3551321029663086, | |
| "learning_rate": 0.000172972972972973, | |
| "loss": 0.6683, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5705705705705706, | |
| "grad_norm": 1.878329873085022, | |
| "learning_rate": 0.00017147147147147148, | |
| "loss": 0.6806, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6006006006006006, | |
| "grad_norm": 1.7099618911743164, | |
| "learning_rate": 0.00016996996996997, | |
| "loss": 0.7677, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6006006006006006, | |
| "eval_accuracy": 0.8533475026567482, | |
| "eval_loss": 0.6426669359207153, | |
| "eval_runtime": 37.3128, | |
| "eval_samples_per_second": 25.219, | |
| "eval_steps_per_second": 3.162, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6306306306306306, | |
| "grad_norm": 3.3745574951171875, | |
| "learning_rate": 0.00016846846846846846, | |
| "loss": 0.7184, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6606606606606606, | |
| "grad_norm": 2.9804413318634033, | |
| "learning_rate": 0.00016696696696696697, | |
| "loss": 0.6027, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6906906906906907, | |
| "grad_norm": 4.826656341552734, | |
| "learning_rate": 0.00016546546546546546, | |
| "loss": 0.6332, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7207207207207207, | |
| "grad_norm": 3.639033317565918, | |
| "learning_rate": 0.00016396396396396395, | |
| "loss": 0.5399, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7507507507507507, | |
| "grad_norm": 3.754734754562378, | |
| "learning_rate": 0.00016246246246246247, | |
| "loss": 0.6603, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7807807807807807, | |
| "grad_norm": 5.434034824371338, | |
| "learning_rate": 0.00016096096096096096, | |
| "loss": 0.6835, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 3.459984302520752, | |
| "learning_rate": 0.00015945945945945947, | |
| "loss": 0.6667, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8408408408408409, | |
| "grad_norm": 4.138125896453857, | |
| "learning_rate": 0.00015795795795795796, | |
| "loss": 0.7526, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8708708708708709, | |
| "grad_norm": 5.941434383392334, | |
| "learning_rate": 0.00015645645645645645, | |
| "loss": 0.5812, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9009009009009009, | |
| "grad_norm": 2.962730884552002, | |
| "learning_rate": 0.00015495495495495496, | |
| "loss": 0.7808, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9009009009009009, | |
| "eval_accuracy": 0.8533475026567482, | |
| "eval_loss": 0.5789640545845032, | |
| "eval_runtime": 37.747, | |
| "eval_samples_per_second": 24.929, | |
| "eval_steps_per_second": 3.126, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9309309309309309, | |
| "grad_norm": 2.512910842895508, | |
| "learning_rate": 0.00015345345345345345, | |
| "loss": 0.7152, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.960960960960961, | |
| "grad_norm": 8.056396484375, | |
| "learning_rate": 0.00015195195195195194, | |
| "loss": 0.5482, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.990990990990991, | |
| "grad_norm": 3.4259350299835205, | |
| "learning_rate": 0.00015045045045045046, | |
| "loss": 0.576, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.021021021021021, | |
| "grad_norm": 4.245713233947754, | |
| "learning_rate": 0.00014894894894894895, | |
| "loss": 0.5693, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0510510510510511, | |
| "grad_norm": 4.1803364753723145, | |
| "learning_rate": 0.00014744744744744746, | |
| "loss": 0.3934, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 3.6080806255340576, | |
| "learning_rate": 0.00014594594594594595, | |
| "loss": 0.598, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 2.7786924839019775, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.5388, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1411411411411412, | |
| "grad_norm": 2.4446325302124023, | |
| "learning_rate": 0.00014294294294294295, | |
| "loss": 0.4524, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1711711711711712, | |
| "grad_norm": 3.9698562622070312, | |
| "learning_rate": 0.00014144144144144144, | |
| "loss": 0.3691, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2012012012012012, | |
| "grad_norm": 2.3817429542541504, | |
| "learning_rate": 0.00013993993993993996, | |
| "loss": 0.3628, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2012012012012012, | |
| "eval_accuracy": 0.8629117959617428, | |
| "eval_loss": 0.5050598382949829, | |
| "eval_runtime": 29.6553, | |
| "eval_samples_per_second": 31.731, | |
| "eval_steps_per_second": 3.979, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2312312312312312, | |
| "grad_norm": 2.1985225677490234, | |
| "learning_rate": 0.00013843843843843845, | |
| "loss": 0.3939, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2612612612612613, | |
| "grad_norm": 2.645564079284668, | |
| "learning_rate": 0.00013693693693693693, | |
| "loss": 0.5042, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2912912912912913, | |
| "grad_norm": 3.640872001647949, | |
| "learning_rate": 0.00013543543543543545, | |
| "loss": 0.6133, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3213213213213213, | |
| "grad_norm": 1.3261995315551758, | |
| "learning_rate": 0.00013393393393393394, | |
| "loss": 0.491, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 5.754721641540527, | |
| "learning_rate": 0.00013243243243243243, | |
| "loss": 0.3445, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3813813813813813, | |
| "grad_norm": 2.1101417541503906, | |
| "learning_rate": 0.00013093093093093094, | |
| "loss": 0.4963, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4114114114114114, | |
| "grad_norm": 2.6475322246551514, | |
| "learning_rate": 0.00012942942942942943, | |
| "loss": 0.4096, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4414414414414414, | |
| "grad_norm": 3.6123669147491455, | |
| "learning_rate": 0.00012792792792792795, | |
| "loss": 0.4881, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4714714714714714, | |
| "grad_norm": 4.201231002807617, | |
| "learning_rate": 0.00012642642642642644, | |
| "loss": 0.609, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5015015015015014, | |
| "grad_norm": 1.5937566757202148, | |
| "learning_rate": 0.00012492492492492492, | |
| "loss": 0.2928, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5015015015015014, | |
| "eval_accuracy": 0.9086078639744952, | |
| "eval_loss": 0.38151538372039795, | |
| "eval_runtime": 25.5673, | |
| "eval_samples_per_second": 36.805, | |
| "eval_steps_per_second": 4.615, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5315315315315314, | |
| "grad_norm": 1.8895208835601807, | |
| "learning_rate": 0.00012342342342342344, | |
| "loss": 0.3181, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5615615615615615, | |
| "grad_norm": 1.6109514236450195, | |
| "learning_rate": 0.00012192192192192193, | |
| "loss": 0.3723, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5915915915915915, | |
| "grad_norm": 1.820304036140442, | |
| "learning_rate": 0.00012042042042042043, | |
| "loss": 0.3654, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 6.402036666870117, | |
| "learning_rate": 0.00011891891891891893, | |
| "loss": 0.4514, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.6516516516516515, | |
| "grad_norm": 4.21787166595459, | |
| "learning_rate": 0.00011741741741741743, | |
| "loss": 0.429, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6816816816816815, | |
| "grad_norm": 6.462326526641846, | |
| "learning_rate": 0.00011591591591591592, | |
| "loss": 0.4445, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7117117117117115, | |
| "grad_norm": 2.5462887287139893, | |
| "learning_rate": 0.00011441441441441443, | |
| "loss": 0.4351, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7417417417417418, | |
| "grad_norm": 6.54508638381958, | |
| "learning_rate": 0.00011291291291291293, | |
| "loss": 0.418, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.7717717717717718, | |
| "grad_norm": 1.9574716091156006, | |
| "learning_rate": 0.00011141141141141143, | |
| "loss": 0.2675, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "grad_norm": 1.2388685941696167, | |
| "learning_rate": 0.00010990990990990993, | |
| "loss": 0.3293, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "eval_accuracy": 0.9064824654622742, | |
| "eval_loss": 0.35215553641319275, | |
| "eval_runtime": 25.5721, | |
| "eval_samples_per_second": 36.798, | |
| "eval_steps_per_second": 4.614, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8318318318318318, | |
| "grad_norm": 13.811619758605957, | |
| "learning_rate": 0.00010840840840840842, | |
| "loss": 0.3406, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.8618618618618619, | |
| "grad_norm": 2.766913890838623, | |
| "learning_rate": 0.00010690690690690692, | |
| "loss": 0.4522, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 0.6912053823471069, | |
| "learning_rate": 0.0001054054054054054, | |
| "loss": 0.5048, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.921921921921922, | |
| "grad_norm": 2.1594743728637695, | |
| "learning_rate": 0.0001039039039039039, | |
| "loss": 0.3974, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.951951951951952, | |
| "grad_norm": 1.6077780723571777, | |
| "learning_rate": 0.0001024024024024024, | |
| "loss": 0.3508, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.981981981981982, | |
| "grad_norm": 5.270846843719482, | |
| "learning_rate": 0.00010090090090090089, | |
| "loss": 0.2338, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.012012012012012, | |
| "grad_norm": 7.626568794250488, | |
| "learning_rate": 9.93993993993994e-05, | |
| "loss": 0.3983, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.042042042042042, | |
| "grad_norm": 1.1183396577835083, | |
| "learning_rate": 9.789789789789791e-05, | |
| "loss": 0.2965, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.0720720720720722, | |
| "grad_norm": 5.2083001136779785, | |
| "learning_rate": 9.639639639639641e-05, | |
| "loss": 0.2886, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.1021021021021022, | |
| "grad_norm": 2.344709873199463, | |
| "learning_rate": 9.48948948948949e-05, | |
| "loss": 0.2239, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1021021021021022, | |
| "eval_accuracy": 0.9086078639744952, | |
| "eval_loss": 0.33202850818634033, | |
| "eval_runtime": 24.5184, | |
| "eval_samples_per_second": 38.379, | |
| "eval_steps_per_second": 4.813, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1321321321321323, | |
| "grad_norm": 7.250065803527832, | |
| "learning_rate": 9.33933933933934e-05, | |
| "loss": 0.3461, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 1.9027929306030273, | |
| "learning_rate": 9.18918918918919e-05, | |
| "loss": 0.2441, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.1921921921921923, | |
| "grad_norm": 3.956188678741455, | |
| "learning_rate": 9.039039039039039e-05, | |
| "loss": 0.2611, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.37913307547569275, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.2862, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.2522522522522523, | |
| "grad_norm": 1.5578209161758423, | |
| "learning_rate": 8.738738738738738e-05, | |
| "loss": 0.2616, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.2822822822822824, | |
| "grad_norm": 1.633107304573059, | |
| "learning_rate": 8.588588588588588e-05, | |
| "loss": 0.2253, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3123123123123124, | |
| "grad_norm": 0.9739989042282104, | |
| "learning_rate": 8.438438438438439e-05, | |
| "loss": 0.2237, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.3423423423423424, | |
| "grad_norm": 1.6076217889785767, | |
| "learning_rate": 8.288288288288289e-05, | |
| "loss": 0.2849, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.3723723723723724, | |
| "grad_norm": 3.5912880897521973, | |
| "learning_rate": 8.138138138138138e-05, | |
| "loss": 0.171, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.4024024024024024, | |
| "grad_norm": 3.782872438430786, | |
| "learning_rate": 7.987987987987988e-05, | |
| "loss": 0.288, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4024024024024024, | |
| "eval_accuracy": 0.9064824654622742, | |
| "eval_loss": 0.35204941034317017, | |
| "eval_runtime": 25.4507, | |
| "eval_samples_per_second": 36.973, | |
| "eval_steps_per_second": 4.636, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 3.8259458541870117, | |
| "learning_rate": 7.837837837837838e-05, | |
| "loss": 0.365, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.4624624624624625, | |
| "grad_norm": 5.670553207397461, | |
| "learning_rate": 7.687687687687688e-05, | |
| "loss": 0.2688, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.4924924924924925, | |
| "grad_norm": 3.157022476196289, | |
| "learning_rate": 7.537537537537538e-05, | |
| "loss": 0.2519, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5225225225225225, | |
| "grad_norm": 6.805280685424805, | |
| "learning_rate": 7.387387387387387e-05, | |
| "loss": 0.4178, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.5525525525525525, | |
| "grad_norm": 2.1328125, | |
| "learning_rate": 7.237237237237238e-05, | |
| "loss": 0.3694, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5825825825825826, | |
| "grad_norm": 4.635671615600586, | |
| "learning_rate": 7.087087087087088e-05, | |
| "loss": 0.2357, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.6126126126126126, | |
| "grad_norm": 4.620871543884277, | |
| "learning_rate": 6.936936936936938e-05, | |
| "loss": 0.2413, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.6426426426426426, | |
| "grad_norm": 0.6175544261932373, | |
| "learning_rate": 6.786786786786787e-05, | |
| "loss": 0.2598, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.6726726726726726, | |
| "grad_norm": 2.0328073501586914, | |
| "learning_rate": 6.636636636636637e-05, | |
| "loss": 0.1978, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 3.2094199657440186, | |
| "learning_rate": 6.486486486486487e-05, | |
| "loss": 0.3209, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_accuracy": 0.9298618490967057, | |
| "eval_loss": 0.28415459394454956, | |
| "eval_runtime": 25.5786, | |
| "eval_samples_per_second": 36.789, | |
| "eval_steps_per_second": 4.613, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.7327327327327327, | |
| "grad_norm": 3.179689407348633, | |
| "learning_rate": 6.336336336336337e-05, | |
| "loss": 0.2128, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.7627627627627627, | |
| "grad_norm": 13.02226734161377, | |
| "learning_rate": 6.186186186186186e-05, | |
| "loss": 0.2989, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.7927927927927927, | |
| "grad_norm": 1.9046058654785156, | |
| "learning_rate": 6.0360360360360365e-05, | |
| "loss": 0.2631, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8228228228228227, | |
| "grad_norm": 0.7185707092285156, | |
| "learning_rate": 5.8858858858858854e-05, | |
| "loss": 0.376, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.8528528528528527, | |
| "grad_norm": 1.6785024404525757, | |
| "learning_rate": 5.7357357357357356e-05, | |
| "loss": 0.1835, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.8828828828828827, | |
| "grad_norm": 10.039284706115723, | |
| "learning_rate": 5.585585585585585e-05, | |
| "loss": 0.2633, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.9129129129129128, | |
| "grad_norm": 1.1975735425949097, | |
| "learning_rate": 5.435435435435435e-05, | |
| "loss": 0.1951, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.942942942942943, | |
| "grad_norm": 4.678313255310059, | |
| "learning_rate": 5.2852852852852855e-05, | |
| "loss": 0.2675, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 2.6503987312316895, | |
| "learning_rate": 5.135135135135135e-05, | |
| "loss": 0.347, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.003003003003003, | |
| "grad_norm": 1.7791227102279663, | |
| "learning_rate": 4.984984984984985e-05, | |
| "loss": 0.187, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.003003003003003, | |
| "eval_accuracy": 0.9351753453772582, | |
| "eval_loss": 0.2576720714569092, | |
| "eval_runtime": 25.5983, | |
| "eval_samples_per_second": 36.76, | |
| "eval_steps_per_second": 4.61, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.033033033033033, | |
| "grad_norm": 3.620664119720459, | |
| "learning_rate": 4.834834834834835e-05, | |
| "loss": 0.2633, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.063063063063063, | |
| "grad_norm": 2.023883819580078, | |
| "learning_rate": 4.684684684684685e-05, | |
| "loss": 0.2504, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.093093093093093, | |
| "grad_norm": 5.697581768035889, | |
| "learning_rate": 4.5345345345345345e-05, | |
| "loss": 0.1817, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.123123123123123, | |
| "grad_norm": 0.10318872332572937, | |
| "learning_rate": 4.384384384384385e-05, | |
| "loss": 0.1818, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.153153153153153, | |
| "grad_norm": 0.06779313087463379, | |
| "learning_rate": 4.234234234234234e-05, | |
| "loss": 0.1635, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.1831831831831834, | |
| "grad_norm": 0.3237922191619873, | |
| "learning_rate": 4.0840840840840845e-05, | |
| "loss": 0.1596, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.2132132132132134, | |
| "grad_norm": 2.90762996673584, | |
| "learning_rate": 3.933933933933934e-05, | |
| "loss": 0.2405, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.2432432432432434, | |
| "grad_norm": 2.0330655574798584, | |
| "learning_rate": 3.783783783783784e-05, | |
| "loss": 0.1546, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.2732732732732734, | |
| "grad_norm": 5.533896446228027, | |
| "learning_rate": 3.633633633633634e-05, | |
| "loss": 0.1483, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.3033033033033035, | |
| "grad_norm": 2.8955769538879395, | |
| "learning_rate": 3.483483483483483e-05, | |
| "loss": 0.1801, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.3033033033033035, | |
| "eval_accuracy": 0.9341126461211477, | |
| "eval_loss": 0.25110554695129395, | |
| "eval_runtime": 25.5935, | |
| "eval_samples_per_second": 36.767, | |
| "eval_steps_per_second": 4.611, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 3.7111897468566895, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1976, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.3633633633633635, | |
| "grad_norm": 0.15836168825626373, | |
| "learning_rate": 3.183183183183183e-05, | |
| "loss": 0.1893, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.3933933933933935, | |
| "grad_norm": 2.5828213691711426, | |
| "learning_rate": 3.0330330330330332e-05, | |
| "loss": 0.1471, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.4234234234234235, | |
| "grad_norm": 5.785639762878418, | |
| "learning_rate": 2.882882882882883e-05, | |
| "loss": 0.2624, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.4534534534534536, | |
| "grad_norm": 2.8061561584472656, | |
| "learning_rate": 2.732732732732733e-05, | |
| "loss": 0.1979, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.4834834834834836, | |
| "grad_norm": 0.6314980983734131, | |
| "learning_rate": 2.582582582582583e-05, | |
| "loss": 0.156, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.5135135135135136, | |
| "grad_norm": 0.12602469325065613, | |
| "learning_rate": 2.4324324324324327e-05, | |
| "loss": 0.1652, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.5435435435435436, | |
| "grad_norm": 4.004512310028076, | |
| "learning_rate": 2.2822822822822822e-05, | |
| "loss": 0.1625, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.5735735735735736, | |
| "grad_norm": 2.2066681385040283, | |
| "learning_rate": 2.132132132132132e-05, | |
| "loss": 0.2252, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.6036036036036037, | |
| "grad_norm": 0.4736880660057068, | |
| "learning_rate": 1.981981981981982e-05, | |
| "loss": 0.2028, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6036036036036037, | |
| "eval_accuracy": 0.946865037194474, | |
| "eval_loss": 0.2209823578596115, | |
| "eval_runtime": 25.6142, | |
| "eval_samples_per_second": 36.737, | |
| "eval_steps_per_second": 4.607, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6336336336336337, | |
| "grad_norm": 0.08393004536628723, | |
| "learning_rate": 1.831831831831832e-05, | |
| "loss": 0.1519, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.6636636636636637, | |
| "grad_norm": 3.0759263038635254, | |
| "learning_rate": 1.6816816816816817e-05, | |
| "loss": 0.269, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.6936936936936937, | |
| "grad_norm": 4.957241535186768, | |
| "learning_rate": 1.5315315315315316e-05, | |
| "loss": 0.1646, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.7237237237237237, | |
| "grad_norm": 3.5457684993743896, | |
| "learning_rate": 1.3813813813813815e-05, | |
| "loss": 0.1377, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.7537537537537538, | |
| "grad_norm": 0.1940850466489792, | |
| "learning_rate": 1.2312312312312313e-05, | |
| "loss": 0.1259, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.7837837837837838, | |
| "grad_norm": 2.370469093322754, | |
| "learning_rate": 1.0810810810810812e-05, | |
| "loss": 0.1706, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.813813813813814, | |
| "grad_norm": 6.3327226638793945, | |
| "learning_rate": 9.309309309309309e-06, | |
| "loss": 0.2907, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.843843843843844, | |
| "grad_norm": 3.131303548812866, | |
| "learning_rate": 7.807807807807808e-06, | |
| "loss": 0.2681, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.873873873873874, | |
| "grad_norm": 4.7293500900268555, | |
| "learning_rate": 6.306306306306306e-06, | |
| "loss": 0.0901, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.903903903903904, | |
| "grad_norm": 0.08293453603982925, | |
| "learning_rate": 4.804804804804805e-06, | |
| "loss": 0.1564, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.903903903903904, | |
| "eval_accuracy": 0.9458023379383634, | |
| "eval_loss": 0.22059296071529388, | |
| "eval_runtime": 25.0695, | |
| "eval_samples_per_second": 37.536, | |
| "eval_steps_per_second": 4.707, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.933933933933934, | |
| "grad_norm": 0.09198717772960663, | |
| "learning_rate": 3.3033033033033035e-06, | |
| "loss": 0.1065, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 3.963963963963964, | |
| "grad_norm": 1.0777033567428589, | |
| "learning_rate": 1.801801801801802e-06, | |
| "loss": 0.2708, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.993993993993994, | |
| "grad_norm": 0.6637467741966248, | |
| "learning_rate": 3.003003003003003e-07, | |
| "loss": 0.2434, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1332, | |
| "total_flos": 1.651775717862015e+18, | |
| "train_loss": 0.47682606708538067, | |
| "train_runtime": 1244.273, | |
| "train_samples_per_second": 17.128, | |
| "train_steps_per_second": 1.071 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1332, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.651775717862015e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |