| { | |
| "best_global_step": 552, | |
| "best_metric": 0.758873745694174, | |
| "best_model_checkpoint": "abcd\\checkpoint-552", | |
| "epoch": 5.157608695652174, | |
| "eval_steps": 500, | |
| "global_step": 552, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.018115942028985508, | |
| "grad_norm": 13.317688941955566, | |
| "learning_rate": 8.035714285714286e-06, | |
| "loss": 0.8905, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.036231884057971016, | |
| "grad_norm": 12.00970458984375, | |
| "learning_rate": 1.6964285714285715e-05, | |
| "loss": 0.7819, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05434782608695652, | |
| "grad_norm": 14.536447525024414, | |
| "learning_rate": 2.5892857142857148e-05, | |
| "loss": 0.9766, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07246376811594203, | |
| "grad_norm": 15.37159252166748, | |
| "learning_rate": 3.4821428571428574e-05, | |
| "loss": 0.8572, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09057971014492754, | |
| "grad_norm": 20.875459671020508, | |
| "learning_rate": 4.375e-05, | |
| "loss": 1.0996, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "grad_norm": 13.146461486816406, | |
| "learning_rate": 4.969758064516129e-05, | |
| "loss": 0.9429, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12681159420289856, | |
| "grad_norm": 20.685447692871094, | |
| "learning_rate": 4.868951612903226e-05, | |
| "loss": 1.1377, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14492753623188406, | |
| "grad_norm": 16.825111389160156, | |
| "learning_rate": 4.768145161290323e-05, | |
| "loss": 1.0243, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16304347826086957, | |
| "grad_norm": 19.70474624633789, | |
| "learning_rate": 4.6673387096774196e-05, | |
| "loss": 1.1325, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.16847826086956522, | |
| "eval_accuracy": 0.556836902800659, | |
| "eval_loss": 1.0675878524780273, | |
| "eval_runtime": 1853.5174, | |
| "eval_samples_per_second": 7.205, | |
| "eval_steps_per_second": 1.801, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.0126811594202898, | |
| "grad_norm": 11.140310287475586, | |
| "learning_rate": 4.566532258064516e-05, | |
| "loss": 0.8952, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0307971014492754, | |
| "grad_norm": 6.155562877655029, | |
| "learning_rate": 4.465725806451613e-05, | |
| "loss": 0.8204, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.048913043478261, | |
| "grad_norm": 9.62833023071289, | |
| "learning_rate": 4.36491935483871e-05, | |
| "loss": 0.8178, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0670289855072463, | |
| "grad_norm": 14.596573829650879, | |
| "learning_rate": 4.2641129032258064e-05, | |
| "loss": 0.9171, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.085144927536232, | |
| "grad_norm": 14.76526927947998, | |
| "learning_rate": 4.163306451612903e-05, | |
| "loss": 0.781, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1032608695652173, | |
| "grad_norm": 21.03084945678711, | |
| "learning_rate": 4.0625000000000005e-05, | |
| "loss": 0.7955, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1213768115942029, | |
| "grad_norm": 18.31418800354004, | |
| "learning_rate": 3.961693548387097e-05, | |
| "loss": 1.1034, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.1394927536231885, | |
| "grad_norm": 17.097795486450195, | |
| "learning_rate": 3.860887096774194e-05, | |
| "loss": 0.8052, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1576086956521738, | |
| "grad_norm": 15.764641761779785, | |
| "learning_rate": 3.7600806451612906e-05, | |
| "loss": 0.738, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1684782608695652, | |
| "eval_accuracy": 0.549947581249064, | |
| "eval_loss": 1.2091021537780762, | |
| "eval_runtime": 1866.8421, | |
| "eval_samples_per_second": 7.153, | |
| "eval_steps_per_second": 1.789, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.0072463768115942, | |
| "grad_norm": 6.231867790222168, | |
| "learning_rate": 3.659274193548387e-05, | |
| "loss": 0.4987, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0253623188405796, | |
| "grad_norm": 7.453345775604248, | |
| "learning_rate": 3.558467741935484e-05, | |
| "loss": 0.3465, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0434782608695654, | |
| "grad_norm": 21.0439453125, | |
| "learning_rate": 3.457661290322581e-05, | |
| "loss": 0.3425, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.0615942028985508, | |
| "grad_norm": 18.279014587402344, | |
| "learning_rate": 3.3568548387096774e-05, | |
| "loss": 0.6557, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.079710144927536, | |
| "grad_norm": 27.888294219970703, | |
| "learning_rate": 3.256048387096775e-05, | |
| "loss": 0.8808, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.097826086956522, | |
| "grad_norm": 16.571508407592773, | |
| "learning_rate": 3.1552419354838715e-05, | |
| "loss": 0.49, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.1159420289855073, | |
| "grad_norm": 23.84330177307129, | |
| "learning_rate": 3.0544354838709676e-05, | |
| "loss": 0.6079, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.1340579710144927, | |
| "grad_norm": 13.54464340209961, | |
| "learning_rate": 2.9536290322580646e-05, | |
| "loss": 0.4746, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.1521739130434785, | |
| "grad_norm": 22.121822357177734, | |
| "learning_rate": 2.8528225806451613e-05, | |
| "loss": 0.6397, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1684782608695654, | |
| "eval_accuracy": 0.611577055563876, | |
| "eval_loss": 1.2484452724456787, | |
| "eval_runtime": 1864.5566, | |
| "eval_samples_per_second": 7.162, | |
| "eval_steps_per_second": 1.791, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 3.0018115942028984, | |
| "grad_norm": 16.256120681762695, | |
| "learning_rate": 2.7520161290322584e-05, | |
| "loss": 0.473, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.0199275362318843, | |
| "grad_norm": 7.968146324157715, | |
| "learning_rate": 2.651209677419355e-05, | |
| "loss": 0.3852, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0380434782608696, | |
| "grad_norm": 21.890714645385742, | |
| "learning_rate": 2.550403225806452e-05, | |
| "loss": 0.4942, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.056159420289855, | |
| "grad_norm": 14.216273307800293, | |
| "learning_rate": 2.4495967741935485e-05, | |
| "loss": 0.3557, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.074275362318841, | |
| "grad_norm": 2.547912836074829, | |
| "learning_rate": 2.3487903225806455e-05, | |
| "loss": 0.3706, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.092391304347826, | |
| "grad_norm": 16.752574920654297, | |
| "learning_rate": 2.247983870967742e-05, | |
| "loss": 0.444, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.1105072463768115, | |
| "grad_norm": 15.96899127960205, | |
| "learning_rate": 2.147177419354839e-05, | |
| "loss": 0.4921, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.128623188405797, | |
| "grad_norm": 11.133148193359375, | |
| "learning_rate": 2.0463709677419356e-05, | |
| "loss": 0.3874, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.1467391304347827, | |
| "grad_norm": 6.035247802734375, | |
| "learning_rate": 1.9455645161290323e-05, | |
| "loss": 0.5042, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.164855072463768, | |
| "grad_norm": 4.174564361572266, | |
| "learning_rate": 1.844758064516129e-05, | |
| "loss": 0.4395, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.1684782608695654, | |
| "eval_accuracy": 0.642728770405871, | |
| "eval_loss": 1.0139626264572144, | |
| "eval_runtime": 1859.3799, | |
| "eval_samples_per_second": 7.182, | |
| "eval_steps_per_second": 1.796, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 4.0144927536231885, | |
| "grad_norm": 3.096120595932007, | |
| "learning_rate": 1.743951612903226e-05, | |
| "loss": 0.2476, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.032608695652174, | |
| "grad_norm": 17.17482566833496, | |
| "learning_rate": 1.6431451612903225e-05, | |
| "loss": 0.324, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.050724637681159, | |
| "grad_norm": 4.938539505004883, | |
| "learning_rate": 1.5423387096774195e-05, | |
| "loss": 0.2958, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.068840579710145, | |
| "grad_norm": 4.778896808624268, | |
| "learning_rate": 1.4415322580645164e-05, | |
| "loss": 0.284, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.086956521739131, | |
| "grad_norm": 20.58946418762207, | |
| "learning_rate": 1.340725806451613e-05, | |
| "loss": 0.4622, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.105072463768116, | |
| "grad_norm": 0.7577747702598572, | |
| "learning_rate": 1.2399193548387098e-05, | |
| "loss": 0.2173, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.1231884057971016, | |
| "grad_norm": 0.33162328600883484, | |
| "learning_rate": 1.1391129032258065e-05, | |
| "loss": 0.2307, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.141304347826087, | |
| "grad_norm": 4.827192783355713, | |
| "learning_rate": 1.0383064516129034e-05, | |
| "loss": 0.3314, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.159420289855072, | |
| "grad_norm": 2.6278650760650635, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 0.3098, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.168478260869565, | |
| "eval_accuracy": 0.6830163246967201, | |
| "eval_loss": 0.9756622910499573, | |
| "eval_runtime": 1864.2768, | |
| "eval_samples_per_second": 7.163, | |
| "eval_steps_per_second": 1.791, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 5.009057971014493, | |
| "grad_norm": 12.96372127532959, | |
| "learning_rate": 8.366935483870968e-06, | |
| "loss": 0.4201, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.0271739130434785, | |
| "grad_norm": 17.302637100219727, | |
| "learning_rate": 7.358870967741936e-06, | |
| "loss": 0.3379, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.045289855072464, | |
| "grad_norm": 4.884481906890869, | |
| "learning_rate": 6.350806451612904e-06, | |
| "loss": 0.3286, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.063405797101449, | |
| "grad_norm": 18.058940887451172, | |
| "learning_rate": 5.342741935483872e-06, | |
| "loss": 0.258, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.081521739130435, | |
| "grad_norm": 1.203333854675293, | |
| "learning_rate": 4.33467741935484e-06, | |
| "loss": 0.1982, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.09963768115942, | |
| "grad_norm": 21.148109436035156, | |
| "learning_rate": 3.3266129032258062e-06, | |
| "loss": 0.2102, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.117753623188406, | |
| "grad_norm": 1.2070428133010864, | |
| "learning_rate": 2.318548387096774e-06, | |
| "loss": 0.2017, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.135869565217392, | |
| "grad_norm": 1.2308259010314941, | |
| "learning_rate": 1.310483870967742e-06, | |
| "loss": 0.1859, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.153985507246377, | |
| "grad_norm": 22.218280792236328, | |
| "learning_rate": 3.024193548387097e-07, | |
| "loss": 0.1016, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.157608695652174, | |
| "eval_accuracy": 0.758873745694174, | |
| "eval_loss": 0.7178874611854553, | |
| "eval_runtime": 1859.6371, | |
| "eval_samples_per_second": 7.181, | |
| "eval_steps_per_second": 1.796, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 5.157608695652174, | |
| "step": 552, | |
| "total_flos": 1.9302137237930803e+18, | |
| "train_loss": 0.5571505176848259, | |
| "train_runtime": 11844.7659, | |
| "train_samples_per_second": 0.186, | |
| "train_steps_per_second": 0.047 | |
| }, | |
| { | |
| "epoch": 5.157608695652174, | |
| "eval_accuracy": 0.7891051507239497, | |
| "eval_loss": 0.6989445090293884, | |
| "eval_runtime": 1206.5933, | |
| "eval_samples_per_second": 6.983, | |
| "eval_steps_per_second": 1.746, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 5.157608695652174, | |
| "eval_accuracy": 0.78867879652753, | |
| "eval_loss": 0.7002037763595581, | |
| "eval_runtime": 1290.721, | |
| "eval_samples_per_second": 6.515, | |
| "eval_steps_per_second": 1.629, | |
| "step": 552 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 552, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9302137237930803e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |