| { | |
| "best_metric": 0.2451845407485962, | |
| "best_model_checkpoint": "./output/checkpoint-1950", | |
| "epoch": 4.20353982300885, | |
| "eval_steps": 150, | |
| "global_step": 2850, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014749262536873156, | |
| "grad_norm": 5.5810041427612305, | |
| "learning_rate": 4.125e-06, | |
| "loss": 0.6821, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.029498525073746312, | |
| "grad_norm": 5.33672571182251, | |
| "learning_rate": 8.25e-06, | |
| "loss": 0.6731, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04424778761061947, | |
| "grad_norm": 3.1765787601470947, | |
| "learning_rate": 1.2375e-05, | |
| "loss": 0.6863, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.058997050147492625, | |
| "grad_norm": 7.523784637451172, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.5096, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07374631268436578, | |
| "grad_norm": 8.258329391479492, | |
| "learning_rate": 2.0625e-05, | |
| "loss": 0.552, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08849557522123894, | |
| "grad_norm": 4.990320205688477, | |
| "learning_rate": 2.475e-05, | |
| "loss": 0.5211, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.10324483775811209, | |
| "grad_norm": 7.771764278411865, | |
| "learning_rate": 2.8874999999999997e-05, | |
| "loss": 0.4423, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.11799410029498525, | |
| "grad_norm": 4.934092044830322, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.5025, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.13274336283185842, | |
| "grad_norm": 5.5473551750183105, | |
| "learning_rate": 3.7125e-05, | |
| "loss": 0.5095, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.14749262536873156, | |
| "grad_norm": 5.446054458618164, | |
| "learning_rate": 4.125e-05, | |
| "loss": 0.439, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16224188790560473, | |
| "grad_norm": 5.696810245513916, | |
| "learning_rate": 4.12495760935163e-05, | |
| "loss": 0.4491, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.17699115044247787, | |
| "grad_norm": 4.427632808685303, | |
| "learning_rate": 4.1248304391490334e-05, | |
| "loss": 0.3156, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.19174041297935104, | |
| "grad_norm": 3.307356357574463, | |
| "learning_rate": 4.1246184946196796e-05, | |
| "loss": 0.3194, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.20648967551622419, | |
| "grad_norm": 1.1623802185058594, | |
| "learning_rate": 4.124321784475777e-05, | |
| "loss": 0.3677, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.22123893805309736, | |
| "grad_norm": 4.190874099731445, | |
| "learning_rate": 4.123940320913919e-05, | |
| "loss": 0.42, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.22123893805309736, | |
| "eval_loss": 0.36307621002197266, | |
| "eval_runtime": 43.6353, | |
| "eval_samples_per_second": 6.921, | |
| "eval_steps_per_second": 6.921, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2359882005899705, | |
| "grad_norm": 1.6925841569900513, | |
| "learning_rate": 4.123474119614577e-05, | |
| "loss": 0.4331, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.25073746312684364, | |
| "grad_norm": 2.9325969219207764, | |
| "learning_rate": 4.1229231997414614e-05, | |
| "loss": 0.4183, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.26548672566371684, | |
| "grad_norm": 2.1130642890930176, | |
| "learning_rate": 4.1222875839407306e-05, | |
| "loss": 0.3555, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.28023598820059, | |
| "grad_norm": 3.514279365539551, | |
| "learning_rate": 4.121567298340059e-05, | |
| "loss": 0.4134, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2949852507374631, | |
| "grad_norm": 3.123249053955078, | |
| "learning_rate": 4.120762372547569e-05, | |
| "loss": 0.5256, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.30973451327433627, | |
| "grad_norm": 2.3070755004882812, | |
| "learning_rate": 4.119872839650605e-05, | |
| "loss": 0.3679, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.32448377581120946, | |
| "grad_norm": 2.9008662700653076, | |
| "learning_rate": 4.118898736214381e-05, | |
| "loss": 0.3923, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3392330383480826, | |
| "grad_norm": 2.0514321327209473, | |
| "learning_rate": 4.117840102280475e-05, | |
| "loss": 0.3725, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 3.542119264602661, | |
| "learning_rate": 4.116696981365181e-05, | |
| "loss": 0.4415, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.3687315634218289, | |
| "grad_norm": 1.6755036115646362, | |
| "learning_rate": 4.115469420457721e-05, | |
| "loss": 0.2841, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3834808259587021, | |
| "grad_norm": 2.9505436420440674, | |
| "learning_rate": 4.1141574700183186e-05, | |
| "loss": 0.3778, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.39823008849557523, | |
| "grad_norm": 2.982377052307129, | |
| "learning_rate": 4.1127611839761155e-05, | |
| "loss": 0.3422, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.41297935103244837, | |
| "grad_norm": 1.809727668762207, | |
| "learning_rate": 4.111280619726964e-05, | |
| "loss": 0.3351, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.4277286135693215, | |
| "grad_norm": 2.7590482234954834, | |
| "learning_rate": 4.109715838131059e-05, | |
| "loss": 0.3718, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.4424778761061947, | |
| "grad_norm": 1.9859437942504883, | |
| "learning_rate": 4.108066903510445e-05, | |
| "loss": 0.3772, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4424778761061947, | |
| "eval_loss": 0.3334226906299591, | |
| "eval_runtime": 43.4221, | |
| "eval_samples_per_second": 6.955, | |
| "eval_steps_per_second": 6.955, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.45722713864306785, | |
| "grad_norm": 2.9821550846099854, | |
| "learning_rate": 4.106333883646366e-05, | |
| "loss": 0.3968, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.471976401179941, | |
| "grad_norm": 3.6904613971710205, | |
| "learning_rate": 4.104516849776479e-05, | |
| "loss": 0.3127, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.48672566371681414, | |
| "grad_norm": 1.0732004642486572, | |
| "learning_rate": 4.1026158765919306e-05, | |
| "loss": 0.4087, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5014749262536873, | |
| "grad_norm": 1.7305632829666138, | |
| "learning_rate": 4.100631042234283e-05, | |
| "loss": 0.4596, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5162241887905604, | |
| "grad_norm": 2.5850343704223633, | |
| "learning_rate": 4.098562428292304e-05, | |
| "loss": 0.3444, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5309734513274337, | |
| "grad_norm": 2.5205276012420654, | |
| "learning_rate": 4.096410119798607e-05, | |
| "loss": 0.4583, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5457227138643068, | |
| "grad_norm": 2.062127113342285, | |
| "learning_rate": 4.094174205226167e-05, | |
| "loss": 0.4003, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.56047197640118, | |
| "grad_norm": 3.1239662170410156, | |
| "learning_rate": 4.0918547764846736e-05, | |
| "loss": 0.3674, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.5752212389380531, | |
| "grad_norm": 2.7132530212402344, | |
| "learning_rate": 4.089451928916758e-05, | |
| "loss": 0.3639, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.5899705014749262, | |
| "grad_norm": 3.0152759552001953, | |
| "learning_rate": 4.0869657612940723e-05, | |
| "loss": 0.2698, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6047197640117994, | |
| "grad_norm": 2.3073341846466064, | |
| "learning_rate": 4.08439637581323e-05, | |
| "loss": 0.439, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.6194690265486725, | |
| "grad_norm": 1.7864807844161987, | |
| "learning_rate": 4.081743878091604e-05, | |
| "loss": 0.2919, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6342182890855457, | |
| "grad_norm": 2.861272096633911, | |
| "learning_rate": 4.079008377162988e-05, | |
| "loss": 0.4066, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.6489675516224189, | |
| "grad_norm": 1.9505175352096558, | |
| "learning_rate": 4.0761899854731085e-05, | |
| "loss": 0.4823, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.6637168141592921, | |
| "grad_norm": 1.8906564712524414, | |
| "learning_rate": 4.073288818875011e-05, | |
| "loss": 0.3265, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6637168141592921, | |
| "eval_loss": 0.30214831233024597, | |
| "eval_runtime": 43.7717, | |
| "eval_samples_per_second": 6.899, | |
| "eval_steps_per_second": 6.899, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6784660766961652, | |
| "grad_norm": 2.3650407791137695, | |
| "learning_rate": 4.070304996624291e-05, | |
| "loss": 0.5034, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.6932153392330384, | |
| "grad_norm": 1.9324402809143066, | |
| "learning_rate": 4.067238641374194e-05, | |
| "loss": 0.349, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 2.0116679668426514, | |
| "learning_rate": 4.0640898791705745e-05, | |
| "loss": 0.5409, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7227138643067846, | |
| "grad_norm": 1.6090418100357056, | |
| "learning_rate": 4.060858839446713e-05, | |
| "loss": 0.3821, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.7374631268436578, | |
| "grad_norm": 2.2912044525146484, | |
| "learning_rate": 4.057545655017998e-05, | |
| "loss": 0.2578, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7522123893805309, | |
| "grad_norm": 1.67880117893219, | |
| "learning_rate": 4.054150462076465e-05, | |
| "loss": 0.3137, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.7669616519174042, | |
| "grad_norm": 1.5085221529006958, | |
| "learning_rate": 4.0506734001851976e-05, | |
| "loss": 0.2617, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.7817109144542773, | |
| "grad_norm": 1.5776855945587158, | |
| "learning_rate": 4.0471146122725904e-05, | |
| "loss": 0.3693, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.7964601769911505, | |
| "grad_norm": 2.036801815032959, | |
| "learning_rate": 4.043474244626477e-05, | |
| "loss": 0.2956, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.8112094395280236, | |
| "grad_norm": 2.230562686920166, | |
| "learning_rate": 4.0397524468881125e-05, | |
| "loss": 0.3842, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8259587020648967, | |
| "grad_norm": 1.4315614700317383, | |
| "learning_rate": 4.0359493720460244e-05, | |
| "loss": 0.3418, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.8407079646017699, | |
| "grad_norm": 3.4652326107025146, | |
| "learning_rate": 4.032065176429724e-05, | |
| "loss": 0.3102, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.855457227138643, | |
| "grad_norm": 1.9565349817276, | |
| "learning_rate": 4.0281000197032795e-05, | |
| "loss": 0.186, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.8702064896755162, | |
| "grad_norm": 2.6647768020629883, | |
| "learning_rate": 4.0240540648587546e-05, | |
| "loss": 0.4584, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.8849557522123894, | |
| "grad_norm": 1.2911219596862793, | |
| "learning_rate": 4.019927478209504e-05, | |
| "loss": 0.2314, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8849557522123894, | |
| "eval_loss": 0.29636457562446594, | |
| "eval_runtime": 43.2686, | |
| "eval_samples_per_second": 6.98, | |
| "eval_steps_per_second": 6.98, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8997050147492626, | |
| "grad_norm": 1.422271728515625, | |
| "learning_rate": 4.015720429383344e-05, | |
| "loss": 0.226, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.9144542772861357, | |
| "grad_norm": 2.965221881866455, | |
| "learning_rate": 4.0114330913155726e-05, | |
| "loss": 0.4821, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.9292035398230089, | |
| "grad_norm": 4.0180158615112305, | |
| "learning_rate": 4.007065640241867e-05, | |
| "loss": 0.2919, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.943952802359882, | |
| "grad_norm": 4.945089817047119, | |
| "learning_rate": 4.002618255691033e-05, | |
| "loss": 0.3667, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.9587020648967551, | |
| "grad_norm": 1.7440930604934692, | |
| "learning_rate": 3.9980911204776306e-05, | |
| "loss": 0.3945, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.9734513274336283, | |
| "grad_norm": 1.846238374710083, | |
| "learning_rate": 3.993484420694458e-05, | |
| "loss": 0.3624, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.9882005899705014, | |
| "grad_norm": 2.105642795562744, | |
| "learning_rate": 3.988798345704899e-05, | |
| "loss": 0.3291, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.0029498525073746, | |
| "grad_norm": 0.9437566995620728, | |
| "learning_rate": 3.984033088135143e-05, | |
| "loss": 0.2252, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.0176991150442478, | |
| "grad_norm": 2.24537992477417, | |
| "learning_rate": 3.979188843866263e-05, | |
| "loss": 0.2462, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.0324483775811208, | |
| "grad_norm": 1.486444354057312, | |
| "learning_rate": 3.97426581202617e-05, | |
| "loss": 0.2597, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.047197640117994, | |
| "grad_norm": 3.2096400260925293, | |
| "learning_rate": 3.969264194981418e-05, | |
| "loss": 0.2238, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.0619469026548674, | |
| "grad_norm": 2.7134766578674316, | |
| "learning_rate": 3.9641841983288953e-05, | |
| "loss": 0.2689, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.0766961651917404, | |
| "grad_norm": 0.9915286898612976, | |
| "learning_rate": 3.959026030887367e-05, | |
| "loss": 0.2326, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.0914454277286136, | |
| "grad_norm": 2.3304569721221924, | |
| "learning_rate": 3.953789904688893e-05, | |
| "loss": 0.2508, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.1061946902654867, | |
| "grad_norm": 0.8963961005210876, | |
| "learning_rate": 3.948476034970113e-05, | |
| "loss": 0.165, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.1061946902654867, | |
| "eval_loss": 0.29341065883636475, | |
| "eval_runtime": 43.6734, | |
| "eval_samples_per_second": 6.915, | |
| "eval_steps_per_second": 6.915, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.12094395280236, | |
| "grad_norm": 1.2763745784759521, | |
| "learning_rate": 3.943084640163398e-05, | |
| "loss": 0.2356, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.135693215339233, | |
| "grad_norm": 1.333479642868042, | |
| "learning_rate": 3.937615941887873e-05, | |
| "loss": 0.2668, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.1504424778761062, | |
| "grad_norm": 2.042940378189087, | |
| "learning_rate": 3.932070164940304e-05, | |
| "loss": 0.2435, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.1651917404129795, | |
| "grad_norm": 2.6705069541931152, | |
| "learning_rate": 3.926447537285859e-05, | |
| "loss": 0.1938, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.1799410029498525, | |
| "grad_norm": 2.413623571395874, | |
| "learning_rate": 3.920748290048739e-05, | |
| "loss": 0.1981, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.1946902654867257, | |
| "grad_norm": 1.9013822078704834, | |
| "learning_rate": 3.914972657502677e-05, | |
| "loss": 0.3461, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.2094395280235988, | |
| "grad_norm": 2.037879228591919, | |
| "learning_rate": 3.9091208770613036e-05, | |
| "loss": 0.2506, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.224188790560472, | |
| "grad_norm": 1.8921838998794556, | |
| "learning_rate": 3.9031931892683937e-05, | |
| "loss": 0.2074, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.238938053097345, | |
| "grad_norm": 1.2936807870864868, | |
| "learning_rate": 3.897189837787975e-05, | |
| "loss": 0.2762, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.2536873156342183, | |
| "grad_norm": 1.600098967552185, | |
| "learning_rate": 3.891111069394313e-05, | |
| "loss": 0.2381, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.2684365781710913, | |
| "grad_norm": 1.6837131977081299, | |
| "learning_rate": 3.884957133961768e-05, | |
| "loss": 0.1811, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.2831858407079646, | |
| "grad_norm": 4.942287921905518, | |
| "learning_rate": 3.878728284454522e-05, | |
| "loss": 0.2511, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.2979351032448379, | |
| "grad_norm": 1.305611491203308, | |
| "learning_rate": 3.872424776916183e-05, | |
| "loss": 0.2289, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.3126843657817109, | |
| "grad_norm": 2.5036911964416504, | |
| "learning_rate": 3.866046870459253e-05, | |
| "loss": 0.4063, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.3274336283185841, | |
| "grad_norm": 2.1621756553649902, | |
| "learning_rate": 3.8595948272544905e-05, | |
| "loss": 0.2515, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3274336283185841, | |
| "eval_loss": 0.278595894575119, | |
| "eval_runtime": 43.7764, | |
| "eval_samples_per_second": 6.899, | |
| "eval_steps_per_second": 6.899, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3421828908554572, | |
| "grad_norm": 0.7632271647453308, | |
| "learning_rate": 3.8530689125201184e-05, | |
| "loss": 0.1884, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.3569321533923304, | |
| "grad_norm": 1.29710054397583, | |
| "learning_rate": 3.8464693945109305e-05, | |
| "loss": 0.225, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.3716814159292037, | |
| "grad_norm": 2.357658863067627, | |
| "learning_rate": 3.839796544507265e-05, | |
| "loss": 0.3185, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.3864306784660767, | |
| "grad_norm": 1.8373112678527832, | |
| "learning_rate": 3.833050636803849e-05, | |
| "loss": 0.29, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.4011799410029497, | |
| "grad_norm": 1.9975624084472656, | |
| "learning_rate": 3.826231948698527e-05, | |
| "loss": 0.3203, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.415929203539823, | |
| "grad_norm": 3.1040427684783936, | |
| "learning_rate": 3.819340760480859e-05, | |
| "loss": 0.2454, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.4306784660766962, | |
| "grad_norm": 1.599753737449646, | |
| "learning_rate": 3.812377355420602e-05, | |
| "loss": 0.2825, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.4454277286135693, | |
| "grad_norm": 1.1874561309814453, | |
| "learning_rate": 3.805342019756065e-05, | |
| "loss": 0.1932, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.4601769911504425, | |
| "grad_norm": 1.8277095556259155, | |
| "learning_rate": 3.7982350426823406e-05, | |
| "loss": 0.2014, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.4749262536873156, | |
| "grad_norm": 0.8104329109191895, | |
| "learning_rate": 3.791056716339421e-05, | |
| "loss": 0.2486, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4896755162241888, | |
| "grad_norm": 2.973177194595337, | |
| "learning_rate": 3.783807335800187e-05, | |
| "loss": 0.2373, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.504424778761062, | |
| "grad_norm": 1.6856945753097534, | |
| "learning_rate": 3.776487199058277e-05, | |
| "loss": 0.2203, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.519174041297935, | |
| "grad_norm": 2.4095230102539062, | |
| "learning_rate": 3.769096607015843e-05, | |
| "loss": 0.2813, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.5339233038348081, | |
| "grad_norm": 2.658792495727539, | |
| "learning_rate": 3.761635863471175e-05, | |
| "loss": 0.2552, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.5486725663716814, | |
| "grad_norm": 2.120602607727051, | |
| "learning_rate": 3.754105275106222e-05, | |
| "loss": 0.3001, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.5486725663716814, | |
| "eval_loss": 0.2679256796836853, | |
| "eval_runtime": 43.6015, | |
| "eval_samples_per_second": 6.926, | |
| "eval_steps_per_second": 6.926, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.5634218289085546, | |
| "grad_norm": 2.71559476852417, | |
| "learning_rate": 3.746505151473972e-05, | |
| "loss": 0.3008, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.5781710914454279, | |
| "grad_norm": 2.8023104667663574, | |
| "learning_rate": 3.738835804985743e-05, | |
| "loss": 0.215, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.592920353982301, | |
| "grad_norm": 1.7893186807632446, | |
| "learning_rate": 3.731097550898329e-05, | |
| "loss": 0.1894, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.607669616519174, | |
| "grad_norm": 1.8876107931137085, | |
| "learning_rate": 3.723290707301047e-05, | |
| "loss": 0.1824, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.6224188790560472, | |
| "grad_norm": 1.5161499977111816, | |
| "learning_rate": 3.7154155951026605e-05, | |
| "loss": 0.192, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6371681415929205, | |
| "grad_norm": 2.110395908355713, | |
| "learning_rate": 3.707472538018187e-05, | |
| "loss": 0.2047, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.6519174041297935, | |
| "grad_norm": 2.1229217052459717, | |
| "learning_rate": 3.6994618625555925e-05, | |
| "loss": 0.1545, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 2.7134130001068115, | |
| "learning_rate": 3.691383898002368e-05, | |
| "loss": 0.2392, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.6814159292035398, | |
| "grad_norm": 1.5704232454299927, | |
| "learning_rate": 3.683238976412e-05, | |
| "loss": 0.1984, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.696165191740413, | |
| "grad_norm": 1.9206628799438477, | |
| "learning_rate": 3.675027432590312e-05, | |
| "loss": 0.2669, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.7109144542772863, | |
| "grad_norm": 2.9219446182250977, | |
| "learning_rate": 3.666749604081707e-05, | |
| "loss": 0.1978, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.7256637168141593, | |
| "grad_norm": 0.8861755728721619, | |
| "learning_rate": 3.6584058311552954e-05, | |
| "loss": 0.1588, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.7404129793510323, | |
| "grad_norm": 3.052067279815674, | |
| "learning_rate": 3.6499964567909e-05, | |
| "loss": 0.1948, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.7551622418879056, | |
| "grad_norm": 2.0168213844299316, | |
| "learning_rate": 3.641521826664964e-05, | |
| "loss": 0.2793, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "grad_norm": 3.0683932304382324, | |
| "learning_rate": 3.63298228913634e-05, | |
| "loss": 0.2384, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "eval_loss": 0.25286030769348145, | |
| "eval_runtime": 43.7965, | |
| "eval_samples_per_second": 6.896, | |
| "eval_steps_per_second": 6.896, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.7846607669616519, | |
| "grad_norm": 2.649764060974121, | |
| "learning_rate": 3.624378195231967e-05, | |
| "loss": 0.3089, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.799410029498525, | |
| "grad_norm": 1.9995458126068115, | |
| "learning_rate": 3.615709898632448e-05, | |
| "loss": 0.2291, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.8141592920353982, | |
| "grad_norm": 2.075753927230835, | |
| "learning_rate": 3.606977755657502e-05, | |
| "loss": 0.2188, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.8289085545722714, | |
| "grad_norm": 3.202075958251953, | |
| "learning_rate": 3.5981821252513274e-05, | |
| "loss": 0.3073, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.8436578171091447, | |
| "grad_norm": 2.0560741424560547, | |
| "learning_rate": 3.5893233689678384e-05, | |
| "loss": 0.2288, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.8584070796460177, | |
| "grad_norm": 2.9390640258789062, | |
| "learning_rate": 3.5804018509558095e-05, | |
| "loss": 0.3001, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.8731563421828907, | |
| "grad_norm": 0.6466512084007263, | |
| "learning_rate": 3.571417937943903e-05, | |
| "loss": 0.1617, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.887905604719764, | |
| "grad_norm": 1.9550355672836304, | |
| "learning_rate": 3.562371999225594e-05, | |
| "loss": 0.2687, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.9026548672566372, | |
| "grad_norm": 2.3754589557647705, | |
| "learning_rate": 3.553264406643995e-05, | |
| "loss": 0.181, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.9174041297935103, | |
| "grad_norm": 1.620802640914917, | |
| "learning_rate": 3.544095534576563e-05, | |
| "loss": 0.2422, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9321533923303835, | |
| "grad_norm": 1.5539398193359375, | |
| "learning_rate": 3.534865759919718e-05, | |
| "loss": 0.1669, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.9469026548672566, | |
| "grad_norm": 2.4959328174591064, | |
| "learning_rate": 3.525575462073344e-05, | |
| "loss": 0.2058, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.9616519174041298, | |
| "grad_norm": 2.10261607170105, | |
| "learning_rate": 3.516225022925199e-05, | |
| "loss": 0.2412, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.976401179941003, | |
| "grad_norm": 2.3935513496398926, | |
| "learning_rate": 3.5068148268352135e-05, | |
| "loss": 0.221, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.991150442477876, | |
| "grad_norm": 1.9170893430709839, | |
| "learning_rate": 3.497345260619691e-05, | |
| "loss": 0.1804, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.991150442477876, | |
| "eval_loss": 0.24877212941646576, | |
| "eval_runtime": 43.1775, | |
| "eval_samples_per_second": 6.994, | |
| "eval_steps_per_second": 6.994, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.005899705014749, | |
| "grad_norm": 1.3050181865692139, | |
| "learning_rate": 3.487816713535409e-05, | |
| "loss": 0.1889, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.0206489675516224, | |
| "grad_norm": 6.0348219871521, | |
| "learning_rate": 3.478229577263617e-05, | |
| "loss": 0.1382, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.0353982300884956, | |
| "grad_norm": 2.732297897338867, | |
| "learning_rate": 3.4685842458939365e-05, | |
| "loss": 0.1052, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.050147492625369, | |
| "grad_norm": 2.3690683841705322, | |
| "learning_rate": 3.458881115908164e-05, | |
| "loss": 0.1049, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 2.0648967551622417, | |
| "grad_norm": 1.0060967206954956, | |
| "learning_rate": 3.449120586163966e-05, | |
| "loss": 0.1413, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.079646017699115, | |
| "grad_norm": 2.224308967590332, | |
| "learning_rate": 3.439303057878493e-05, | |
| "loss": 0.1162, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.094395280235988, | |
| "grad_norm": 1.902258038520813, | |
| "learning_rate": 3.429428934611879e-05, | |
| "loss": 0.1231, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.1091445427728615, | |
| "grad_norm": 3.415437936782837, | |
| "learning_rate": 3.419498622250657e-05, | |
| "loss": 0.1914, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 2.1238938053097347, | |
| "grad_norm": 2.006056070327759, | |
| "learning_rate": 3.409512528991075e-05, | |
| "loss": 0.1494, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.1386430678466075, | |
| "grad_norm": 2.260693311691284, | |
| "learning_rate": 3.399471065322314e-05, | |
| "loss": 0.1251, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.1533923303834808, | |
| "grad_norm": 2.281259298324585, | |
| "learning_rate": 3.3893746440096144e-05, | |
| "loss": 0.1238, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.168141592920354, | |
| "grad_norm": 1.1317875385284424, | |
| "learning_rate": 3.3792236800773114e-05, | |
| "loss": 0.1144, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 2.1828908554572273, | |
| "grad_norm": 2.4440741539001465, | |
| "learning_rate": 3.369018590791776e-05, | |
| "loss": 0.1123, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.1976401179941005, | |
| "grad_norm": 1.8043389320373535, | |
| "learning_rate": 3.358759795644255e-05, | |
| "loss": 0.1347, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 2.2123893805309733, | |
| "grad_norm": 3.1604647636413574, | |
| "learning_rate": 3.3484477163336383e-05, | |
| "loss": 0.1287, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.2123893805309733, | |
| "eval_loss": 0.25868239998817444, | |
| "eval_runtime": 43.8975, | |
| "eval_samples_per_second": 6.88, | |
| "eval_steps_per_second": 6.88, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.2271386430678466, | |
| "grad_norm": 2.021027088165283, | |
| "learning_rate": 3.338082776749115e-05, | |
| "loss": 0.1872, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 2.24188790560472, | |
| "grad_norm": 1.5057893991470337, | |
| "learning_rate": 3.327665402952756e-05, | |
| "loss": 0.1177, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.256637168141593, | |
| "grad_norm": 1.7474180459976196, | |
| "learning_rate": 3.317196023161996e-05, | |
| "loss": 0.1443, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 2.271386430678466, | |
| "grad_norm": 1.8646279573440552, | |
| "learning_rate": 3.306675067732031e-05, | |
| "loss": 0.1386, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.286135693215339, | |
| "grad_norm": 0.44584158062934875, | |
| "learning_rate": 3.296102969138133e-05, | |
| "loss": 0.1612, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.3008849557522124, | |
| "grad_norm": 0.7199454307556152, | |
| "learning_rate": 3.285480161957865e-05, | |
| "loss": 0.1217, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 2.3156342182890857, | |
| "grad_norm": 1.7795122861862183, | |
| "learning_rate": 3.274807082853226e-05, | |
| "loss": 0.1666, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 2.330383480825959, | |
| "grad_norm": 1.9658058881759644, | |
| "learning_rate": 3.264084170552692e-05, | |
| "loss": 0.1036, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.3451327433628317, | |
| "grad_norm": 0.996110200881958, | |
| "learning_rate": 3.25331186583319e-05, | |
| "loss": 0.1381, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 2.359882005899705, | |
| "grad_norm": 5.947177410125732, | |
| "learning_rate": 3.242490611501975e-05, | |
| "loss": 0.1579, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.3746312684365782, | |
| "grad_norm": 3.3651394844055176, | |
| "learning_rate": 3.231620852378428e-05, | |
| "loss": 0.165, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 2.3893805309734515, | |
| "grad_norm": 1.959834098815918, | |
| "learning_rate": 3.220703035275773e-05, | |
| "loss": 0.1379, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 2.4041297935103243, | |
| "grad_norm": 2.7289717197418213, | |
| "learning_rate": 3.209737608982709e-05, | |
| "loss": 0.196, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 2.4188790560471976, | |
| "grad_norm": 4.184900283813477, | |
| "learning_rate": 3.1987250242449616e-05, | |
| "loss": 0.1701, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 2.433628318584071, | |
| "grad_norm": 2.6607248783111572, | |
| "learning_rate": 3.1876657337467564e-05, | |
| "loss": 0.1057, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.433628318584071, | |
| "eval_loss": 0.2500734031200409, | |
| "eval_runtime": 44.0, | |
| "eval_samples_per_second": 6.864, | |
| "eval_steps_per_second": 6.864, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.448377581120944, | |
| "grad_norm": 3.242337465286255, | |
| "learning_rate": 3.176560192092211e-05, | |
| "loss": 0.1478, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 2.4631268436578173, | |
| "grad_norm": 2.5350615978240967, | |
| "learning_rate": 3.165408855786642e-05, | |
| "loss": 0.1988, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 2.47787610619469, | |
| "grad_norm": 2.062612295150757, | |
| "learning_rate": 3.154212183217812e-05, | |
| "loss": 0.1228, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.4926253687315634, | |
| "grad_norm": 2.87460994720459, | |
| "learning_rate": 3.142970634637072e-05, | |
| "loss": 0.1057, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 2.5073746312684366, | |
| "grad_norm": 1.2329528331756592, | |
| "learning_rate": 3.131684672140458e-05, | |
| "loss": 0.1142, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.52212389380531, | |
| "grad_norm": 2.1753523349761963, | |
| "learning_rate": 3.120354759649682e-05, | |
| "loss": 0.1065, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 2.5368731563421827, | |
| "grad_norm": 2.7612011432647705, | |
| "learning_rate": 3.1089813628930695e-05, | |
| "loss": 0.189, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 2.551622418879056, | |
| "grad_norm": 1.8067593574523926, | |
| "learning_rate": 3.097564949386416e-05, | |
| "loss": 0.1085, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 2.566371681415929, | |
| "grad_norm": 2.544847011566162, | |
| "learning_rate": 3.086105988413766e-05, | |
| "loss": 0.2134, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 2.5811209439528024, | |
| "grad_norm": 0.837935209274292, | |
| "learning_rate": 3.074604951008122e-05, | |
| "loss": 0.0964, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.5958702064896757, | |
| "grad_norm": 1.2727034091949463, | |
| "learning_rate": 3.063062309932086e-05, | |
| "loss": 0.1179, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 2.6106194690265485, | |
| "grad_norm": 2.784679651260376, | |
| "learning_rate": 3.0514785396584238e-05, | |
| "loss": 0.1062, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 2.6253687315634218, | |
| "grad_norm": 1.9828710556030273, | |
| "learning_rate": 3.0398541163505598e-05, | |
| "loss": 0.1325, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 2.640117994100295, | |
| "grad_norm": 1.7931956052780151, | |
| "learning_rate": 3.028189517843007e-05, | |
| "loss": 0.1326, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 2.6548672566371683, | |
| "grad_norm": 2.979793071746826, | |
| "learning_rate": 3.0164852236217233e-05, | |
| "loss": 0.1903, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.6548672566371683, | |
| "eval_loss": 0.2538958191871643, | |
| "eval_runtime": 43.8985, | |
| "eval_samples_per_second": 6.879, | |
| "eval_steps_per_second": 6.879, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.669616519174041, | |
| "grad_norm": 1.796676516532898, | |
| "learning_rate": 3.0047417148044e-05, | |
| "loss": 0.125, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 2.6843657817109143, | |
| "grad_norm": 3.8405046463012695, | |
| "learning_rate": 2.99295947412069e-05, | |
| "loss": 0.119, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 2.6991150442477876, | |
| "grad_norm": 2.2203421592712402, | |
| "learning_rate": 2.9811389858923593e-05, | |
| "loss": 0.1104, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 2.713864306784661, | |
| "grad_norm": 1.727965235710144, | |
| "learning_rate": 2.9692807360133822e-05, | |
| "loss": 0.1523, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 2.728613569321534, | |
| "grad_norm": 2.0604889392852783, | |
| "learning_rate": 2.9573852119299634e-05, | |
| "loss": 0.1517, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.7433628318584073, | |
| "grad_norm": 2.538522243499756, | |
| "learning_rate": 2.9454529026205092e-05, | |
| "loss": 0.1528, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 2.75811209439528, | |
| "grad_norm": 1.8873978853225708, | |
| "learning_rate": 2.9334842985755173e-05, | |
| "loss": 0.1234, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 2.7728613569321534, | |
| "grad_norm": 1.9439120292663574, | |
| "learning_rate": 2.921479891777423e-05, | |
| "loss": 0.1629, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 2.7876106194690267, | |
| "grad_norm": 1.4135169982910156, | |
| "learning_rate": 2.9094401756803725e-05, | |
| "loss": 0.1266, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 2.8023598820058995, | |
| "grad_norm": 3.5280425548553467, | |
| "learning_rate": 2.8973656451899372e-05, | |
| "loss": 0.1209, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.8171091445427727, | |
| "grad_norm": 3.0947368144989014, | |
| "learning_rate": 2.8852567966427735e-05, | |
| "loss": 0.1104, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 2.831858407079646, | |
| "grad_norm": 2.14790940284729, | |
| "learning_rate": 2.8731141277862174e-05, | |
| "loss": 0.1489, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 2.8466076696165192, | |
| "grad_norm": 2.191667079925537, | |
| "learning_rate": 2.8609381377578267e-05, | |
| "loss": 0.0923, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 2.8613569321533925, | |
| "grad_norm": 1.2369720935821533, | |
| "learning_rate": 2.848729327064861e-05, | |
| "loss": 0.0907, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 2.8761061946902657, | |
| "grad_norm": 3.589946746826172, | |
| "learning_rate": 2.8364881975637094e-05, | |
| "loss": 0.1721, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.8761061946902657, | |
| "eval_loss": 0.2451845407485962, | |
| "eval_runtime": 43.8481, | |
| "eval_samples_per_second": 6.887, | |
| "eval_steps_per_second": 6.887, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.8908554572271385, | |
| "grad_norm": 2.7611799240112305, | |
| "learning_rate": 2.82421525243926e-05, | |
| "loss": 0.1254, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 2.905604719764012, | |
| "grad_norm": 1.5316367149353027, | |
| "learning_rate": 2.8119109961842176e-05, | |
| "loss": 0.1056, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 2.920353982300885, | |
| "grad_norm": 1.325637698173523, | |
| "learning_rate": 2.799575934578365e-05, | |
| "loss": 0.1512, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 2.935103244837758, | |
| "grad_norm": 2.2634294033050537, | |
| "learning_rate": 2.7872105746677694e-05, | |
| "loss": 0.1763, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 2.949852507374631, | |
| "grad_norm": 1.410277009010315, | |
| "learning_rate": 2.774815424743947e-05, | |
| "loss": 0.2031, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.9646017699115044, | |
| "grad_norm": 3.4232828617095947, | |
| "learning_rate": 2.762390994322962e-05, | |
| "loss": 0.1413, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 2.9793510324483776, | |
| "grad_norm": 1.2643831968307495, | |
| "learning_rate": 2.749937794124486e-05, | |
| "loss": 0.1268, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 2.994100294985251, | |
| "grad_norm": 3.378948926925659, | |
| "learning_rate": 2.7374563360508036e-05, | |
| "loss": 0.1299, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 3.0088495575221237, | |
| "grad_norm": 2.396994113922119, | |
| "learning_rate": 2.7249471331657693e-05, | |
| "loss": 0.0974, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 3.023598820058997, | |
| "grad_norm": 1.482927918434143, | |
| "learning_rate": 2.712410699673718e-05, | |
| "loss": 0.098, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.03834808259587, | |
| "grad_norm": 1.5194566249847412, | |
| "learning_rate": 2.699847550898329e-05, | |
| "loss": 0.0639, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 3.0530973451327434, | |
| "grad_norm": 1.4694433212280273, | |
| "learning_rate": 2.6872582032614426e-05, | |
| "loss": 0.1063, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 3.0678466076696167, | |
| "grad_norm": 0.5857545137405396, | |
| "learning_rate": 2.6746431742618305e-05, | |
| "loss": 0.0709, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 3.0825958702064895, | |
| "grad_norm": 2.3842179775238037, | |
| "learning_rate": 2.6620029824539257e-05, | |
| "loss": 0.1083, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 3.0973451327433628, | |
| "grad_norm": 1.8553909063339233, | |
| "learning_rate": 2.6493381474265044e-05, | |
| "loss": 0.0786, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.0973451327433628, | |
| "eval_loss": 0.2644558846950531, | |
| "eval_runtime": 43.8036, | |
| "eval_samples_per_second": 6.894, | |
| "eval_steps_per_second": 6.894, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.112094395280236, | |
| "grad_norm": 0.7660773992538452, | |
| "learning_rate": 2.636649189781331e-05, | |
| "loss": 0.0781, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 3.1268436578171093, | |
| "grad_norm": 2.0916807651519775, | |
| "learning_rate": 2.6239366311117528e-05, | |
| "loss": 0.129, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 3.1415929203539825, | |
| "grad_norm": 1.9852261543273926, | |
| "learning_rate": 2.6112009939812672e-05, | |
| "loss": 0.0631, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 3.1563421828908553, | |
| "grad_norm": 1.6749731302261353, | |
| "learning_rate": 2.5984428019020343e-05, | |
| "loss": 0.0932, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 3.1710914454277286, | |
| "grad_norm": 2.2517542839050293, | |
| "learning_rate": 2.5856625793133585e-05, | |
| "loss": 0.0752, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 3.185840707964602, | |
| "grad_norm": 3.476151466369629, | |
| "learning_rate": 2.5728608515601357e-05, | |
| "loss": 0.0657, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 3.200589970501475, | |
| "grad_norm": 8.088579177856445, | |
| "learning_rate": 2.560038144871252e-05, | |
| "loss": 0.1153, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 3.215339233038348, | |
| "grad_norm": 2.3388564586639404, | |
| "learning_rate": 2.547194986337956e-05, | |
| "loss": 0.0684, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 3.230088495575221, | |
| "grad_norm": 2.223625421524048, | |
| "learning_rate": 2.5343319038921927e-05, | |
| "loss": 0.0712, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 3.2448377581120944, | |
| "grad_norm": 1.7080127000808716, | |
| "learning_rate": 2.521449426284898e-05, | |
| "loss": 0.0701, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.2595870206489677, | |
| "grad_norm": 1.1411865949630737, | |
| "learning_rate": 2.5085480830642722e-05, | |
| "loss": 0.0773, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 3.274336283185841, | |
| "grad_norm": 3.657041072845459, | |
| "learning_rate": 2.4956284045540015e-05, | |
| "loss": 0.0779, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 3.2890855457227137, | |
| "grad_norm": 1.847219705581665, | |
| "learning_rate": 2.4826909218314684e-05, | |
| "loss": 0.0656, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 3.303834808259587, | |
| "grad_norm": 1.5679715871810913, | |
| "learning_rate": 2.4697361667059132e-05, | |
| "loss": 0.0741, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 3.3185840707964602, | |
| "grad_norm": 2.4843060970306396, | |
| "learning_rate": 2.4567646716965808e-05, | |
| "loss": 0.077, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.3185840707964602, | |
| "eval_loss": 0.2629007399082184, | |
| "eval_runtime": 43.9182, | |
| "eval_samples_per_second": 6.876, | |
| "eval_steps_per_second": 6.876, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 1.421711802482605, | |
| "learning_rate": 2.443776970010823e-05, | |
| "loss": 0.0612, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 3.3480825958702063, | |
| "grad_norm": 1.3979884386062622, | |
| "learning_rate": 2.430773595522188e-05, | |
| "loss": 0.0962, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 3.3628318584070795, | |
| "grad_norm": 2.23537540435791, | |
| "learning_rate": 2.4177550827484704e-05, | |
| "loss": 0.1082, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 3.377581120943953, | |
| "grad_norm": 0.7765734195709229, | |
| "learning_rate": 2.4047219668297402e-05, | |
| "loss": 0.1427, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 3.392330383480826, | |
| "grad_norm": 1.6800991296768188, | |
| "learning_rate": 2.3916747835063446e-05, | |
| "loss": 0.0701, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.4070796460176993, | |
| "grad_norm": 3.2461161613464355, | |
| "learning_rate": 2.3786140690968887e-05, | |
| "loss": 0.0997, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 3.421828908554572, | |
| "grad_norm": 6.6169257164001465, | |
| "learning_rate": 2.3655403604761872e-05, | |
| "loss": 0.0774, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 3.4365781710914454, | |
| "grad_norm": 0.5882118940353394, | |
| "learning_rate": 2.3524541950531934e-05, | |
| "loss": 0.0857, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 3.4513274336283186, | |
| "grad_norm": 1.4368208646774292, | |
| "learning_rate": 2.3393561107489144e-05, | |
| "loss": 0.1152, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 3.466076696165192, | |
| "grad_norm": 1.8174480199813843, | |
| "learning_rate": 2.3262466459742938e-05, | |
| "loss": 0.0788, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 3.4808259587020647, | |
| "grad_norm": 4.503509044647217, | |
| "learning_rate": 2.313126339608082e-05, | |
| "loss": 0.0961, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 3.495575221238938, | |
| "grad_norm": 1.568178415298462, | |
| "learning_rate": 2.2999957309746853e-05, | |
| "loss": 0.0868, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 3.510324483775811, | |
| "grad_norm": 2.2628488540649414, | |
| "learning_rate": 2.286855359821995e-05, | |
| "loss": 0.0766, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 3.5250737463126844, | |
| "grad_norm": 2.249336004257202, | |
| "learning_rate": 2.273705766299202e-05, | |
| "loss": 0.1004, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "grad_norm": 2.4996466636657715, | |
| "learning_rate": 2.2605474909345937e-05, | |
| "loss": 0.0864, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.5398230088495577, | |
| "eval_loss": 0.2607187330722809, | |
| "eval_runtime": 43.5839, | |
| "eval_samples_per_second": 6.929, | |
| "eval_steps_per_second": 6.929, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.554572271386431, | |
| "grad_norm": 2.188922643661499, | |
| "learning_rate": 2.2473810746133318e-05, | |
| "loss": 0.1021, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 3.5693215339233038, | |
| "grad_norm": 1.4103885889053345, | |
| "learning_rate": 2.234207058555222e-05, | |
| "loss": 0.1025, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 3.584070796460177, | |
| "grad_norm": 3.289332389831543, | |
| "learning_rate": 2.221025984292466e-05, | |
| "loss": 0.0835, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 3.5988200589970503, | |
| "grad_norm": 1.7993652820587158, | |
| "learning_rate": 2.2078383936473987e-05, | |
| "loss": 0.0776, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 3.613569321533923, | |
| "grad_norm": 2.0649101734161377, | |
| "learning_rate": 2.1946448287102206e-05, | |
| "loss": 0.0973, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 3.6283185840707963, | |
| "grad_norm": 1.656886339187622, | |
| "learning_rate": 2.18144583181671e-05, | |
| "loss": 0.0863, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 3.6430678466076696, | |
| "grad_norm": 2.353787899017334, | |
| "learning_rate": 2.168241945525932e-05, | |
| "loss": 0.0962, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 3.657817109144543, | |
| "grad_norm": 2.442674398422241, | |
| "learning_rate": 2.1550337125979373e-05, | |
| "loss": 0.0828, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 3.672566371681416, | |
| "grad_norm": 1.9212013483047485, | |
| "learning_rate": 2.1418216759714467e-05, | |
| "loss": 0.0729, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 3.6873156342182893, | |
| "grad_norm": 3.9996650218963623, | |
| "learning_rate": 2.1286063787415392e-05, | |
| "loss": 0.1083, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.702064896755162, | |
| "grad_norm": 1.5336638689041138, | |
| "learning_rate": 2.115388364137322e-05, | |
| "loss": 0.0758, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 3.7168141592920354, | |
| "grad_norm": 0.8267903923988342, | |
| "learning_rate": 2.1021681754996045e-05, | |
| "loss": 0.0739, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 3.7315634218289087, | |
| "grad_norm": 0.9670748114585876, | |
| "learning_rate": 2.0889463562585625e-05, | |
| "loss": 0.085, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 3.7463126843657815, | |
| "grad_norm": 1.723310947418213, | |
| "learning_rate": 2.075723449911398e-05, | |
| "loss": 0.066, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 3.7610619469026547, | |
| "grad_norm": 2.4782888889312744, | |
| "learning_rate": 2.0625e-05, | |
| "loss": 0.0905, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 3.7610619469026547, | |
| "eval_loss": 0.2644849419593811, | |
| "eval_runtime": 43.9466, | |
| "eval_samples_per_second": 6.872, | |
| "eval_steps_per_second": 6.872, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 3.775811209439528, | |
| "grad_norm": 2.117515802383423, | |
| "learning_rate": 2.049276550088602e-05, | |
| "loss": 0.0745, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 3.7905604719764012, | |
| "grad_norm": 2.7794859409332275, | |
| "learning_rate": 2.0360536437414375e-05, | |
| "loss": 0.0939, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 3.8053097345132745, | |
| "grad_norm": 1.0744003057479858, | |
| "learning_rate": 2.0228318245003955e-05, | |
| "loss": 0.0675, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 3.8200589970501477, | |
| "grad_norm": 1.0996911525726318, | |
| "learning_rate": 2.0096116358626783e-05, | |
| "loss": 0.0662, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 3.8348082595870205, | |
| "grad_norm": 0.7364205718040466, | |
| "learning_rate": 1.996393621258461e-05, | |
| "loss": 0.0553, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.849557522123894, | |
| "grad_norm": 1.5476138591766357, | |
| "learning_rate": 1.9831783240285533e-05, | |
| "loss": 0.0942, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 3.864306784660767, | |
| "grad_norm": 1.6006284952163696, | |
| "learning_rate": 1.9699662874020637e-05, | |
| "loss": 0.0969, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 3.87905604719764, | |
| "grad_norm": 0.5195714831352234, | |
| "learning_rate": 1.9567580544740682e-05, | |
| "loss": 0.0875, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 3.893805309734513, | |
| "grad_norm": 0.9594073295593262, | |
| "learning_rate": 1.9435541681832906e-05, | |
| "loss": 0.0713, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 3.9085545722713864, | |
| "grad_norm": 0.8206641674041748, | |
| "learning_rate": 1.9303551712897798e-05, | |
| "loss": 0.1369, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 3.9233038348082596, | |
| "grad_norm": 2.239541530609131, | |
| "learning_rate": 1.9171616063526012e-05, | |
| "loss": 0.0837, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 3.938053097345133, | |
| "grad_norm": 1.1113237142562866, | |
| "learning_rate": 1.9039740157075348e-05, | |
| "loss": 0.0671, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 3.952802359882006, | |
| "grad_norm": 2.035332441329956, | |
| "learning_rate": 1.8907929414447785e-05, | |
| "loss": 0.0955, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 3.967551622418879, | |
| "grad_norm": 0.7038524150848389, | |
| "learning_rate": 1.8776189253866686e-05, | |
| "loss": 0.0708, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 3.982300884955752, | |
| "grad_norm": 3.023179054260254, | |
| "learning_rate": 1.8644525090654063e-05, | |
| "loss": 0.0863, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.982300884955752, | |
| "eval_loss": 0.27085980772972107, | |
| "eval_runtime": 43.594, | |
| "eval_samples_per_second": 6.928, | |
| "eval_steps_per_second": 6.928, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.9970501474926254, | |
| "grad_norm": 0.3948822021484375, | |
| "learning_rate": 1.851294233700798e-05, | |
| "loss": 0.058, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 4.011799410029498, | |
| "grad_norm": 2.1759111881256104, | |
| "learning_rate": 1.8381446401780052e-05, | |
| "loss": 0.0544, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 4.0265486725663715, | |
| "grad_norm": 1.0792875289916992, | |
| "learning_rate": 1.825004269025315e-05, | |
| "loss": 0.0438, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 4.041297935103245, | |
| "grad_norm": 0.6035469770431519, | |
| "learning_rate": 1.811873660391918e-05, | |
| "loss": 0.068, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 4.056047197640118, | |
| "grad_norm": 2.127488613128662, | |
| "learning_rate": 1.7987533540257062e-05, | |
| "loss": 0.0842, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 4.070796460176991, | |
| "grad_norm": 0.9804584980010986, | |
| "learning_rate": 1.7856438892510862e-05, | |
| "loss": 0.0512, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 4.0855457227138645, | |
| "grad_norm": 0.3105282187461853, | |
| "learning_rate": 1.772545804946807e-05, | |
| "loss": 0.0632, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 4.100294985250738, | |
| "grad_norm": 0.9263339638710022, | |
| "learning_rate": 1.759459639523813e-05, | |
| "loss": 0.0495, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 4.115044247787611, | |
| "grad_norm": 0.42594772577285767, | |
| "learning_rate": 1.7463859309031106e-05, | |
| "loss": 0.0467, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 4.129793510324483, | |
| "grad_norm": 1.5321959257125854, | |
| "learning_rate": 1.7333252164936557e-05, | |
| "loss": 0.0622, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.144542772861357, | |
| "grad_norm": 0.47825008630752563, | |
| "learning_rate": 1.7202780331702608e-05, | |
| "loss": 0.0667, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 4.15929203539823, | |
| "grad_norm": 2.095520496368408, | |
| "learning_rate": 1.70724491725153e-05, | |
| "loss": 0.063, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 4.174041297935103, | |
| "grad_norm": 0.4359021782875061, | |
| "learning_rate": 1.694226404477812e-05, | |
| "loss": 0.0429, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 4.188790560471976, | |
| "grad_norm": 2.7264065742492676, | |
| "learning_rate": 1.681223029989177e-05, | |
| "loss": 0.0501, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 4.20353982300885, | |
| "grad_norm": 0.8709071278572083, | |
| "learning_rate": 1.66823532830342e-05, | |
| "loss": 0.0464, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 4.20353982300885, | |
| "eval_loss": 0.29095226526260376, | |
| "eval_runtime": 43.6544, | |
| "eval_samples_per_second": 6.918, | |
| "eval_steps_per_second": 6.918, | |
| "step": 2850 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 150, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.313588061300654e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |