| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 8736, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005723443223443223, | |
| "grad_norm": 6.21875, | |
| "learning_rate": 1.4302059496567508e-06, | |
| "loss": 3.2916, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.011446886446886446, | |
| "grad_norm": 3.453125, | |
| "learning_rate": 2.8604118993135015e-06, | |
| "loss": 3.1002, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.011446886446886446, | |
| "eval_loss": 3.050471782684326, | |
| "eval_runtime": 124.1248, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.017170329670329672, | |
| "grad_norm": 1.8671875, | |
| "learning_rate": 4.290617848970252e-06, | |
| "loss": 2.6977, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.022893772893772892, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 5.720823798627003e-06, | |
| "loss": 2.1929, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.022893772893772892, | |
| "eval_loss": 2.049286365509033, | |
| "eval_runtime": 124.1321, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.028617216117216116, | |
| "grad_norm": 0.625, | |
| "learning_rate": 7.151029748283754e-06, | |
| "loss": 1.8298, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.034340659340659344, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 8.581235697940504e-06, | |
| "loss": 1.6369, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.034340659340659344, | |
| "eval_loss": 1.6432359218597412, | |
| "eval_runtime": 124.1394, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.04006410256410257, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 1.0011441647597253e-05, | |
| "loss": 1.531, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.045787545787545784, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 1.1441647597254006e-05, | |
| "loss": 1.4618, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.045787545787545784, | |
| "eval_loss": 1.5580341815948486, | |
| "eval_runtime": 124.1039, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.05151098901098901, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 1.2871853546910755e-05, | |
| "loss": 1.4061, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.05723443223443223, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 1.4302059496567508e-05, | |
| "loss": 1.317, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05723443223443223, | |
| "eval_loss": 1.5409735441207886, | |
| "eval_runtime": 124.0925, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06295787545787546, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 1.5732265446224257e-05, | |
| "loss": 1.2334, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.06868131868131869, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 1.716247139588101e-05, | |
| "loss": 1.1329, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.06868131868131869, | |
| "eval_loss": 1.6269210577011108, | |
| "eval_runtime": 124.0849, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.0744047619047619, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 1.859267734553776e-05, | |
| "loss": 1.039, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.08012820512820513, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 2.0022883295194507e-05, | |
| "loss": 0.9505, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.08012820512820513, | |
| "eval_loss": 1.738716959953308, | |
| "eval_runtime": 124.1002, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.08585164835164835, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 2.145308924485126e-05, | |
| "loss": 0.8827, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.09157509157509157, | |
| "grad_norm": 0.4296875, | |
| "learning_rate": 2.2883295194508012e-05, | |
| "loss": 0.8334, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.09157509157509157, | |
| "eval_loss": 1.7442790269851685, | |
| "eval_runtime": 124.087, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.0972985347985348, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 2.431350114416476e-05, | |
| "loss": 0.8127, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.10302197802197802, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 2.574370709382151e-05, | |
| "loss": 0.7692, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.10302197802197802, | |
| "eval_loss": 1.7634161710739136, | |
| "eval_runtime": 124.1852, | |
| "eval_samples_per_second": 2.255, | |
| "eval_steps_per_second": 0.564, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.10874542124542125, | |
| "grad_norm": 0.455078125, | |
| "learning_rate": 2.7173913043478262e-05, | |
| "loss": 0.74, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.11446886446886446, | |
| "grad_norm": 0.48828125, | |
| "learning_rate": 2.8604118993135016e-05, | |
| "loss": 0.6983, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11446886446886446, | |
| "eval_loss": 1.7546391487121582, | |
| "eval_runtime": 124.0637, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1201923076923077, | |
| "grad_norm": 0.46484375, | |
| "learning_rate": 3.0034324942791764e-05, | |
| "loss": 0.7277, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.1259157509157509, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 3.1464530892448515e-05, | |
| "loss": 0.6859, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1259157509157509, | |
| "eval_loss": 1.7593410015106201, | |
| "eval_runtime": 124.125, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.13163919413919414, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 3.289473684210527e-05, | |
| "loss": 0.6844, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.13736263736263737, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 3.432494279176202e-05, | |
| "loss": 0.6671, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.13736263736263737, | |
| "eval_loss": 1.7647184133529663, | |
| "eval_runtime": 124.0962, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.14308608058608058, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 3.5755148741418764e-05, | |
| "loss": 0.6553, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.1488095238095238, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 3.718535469107552e-05, | |
| "loss": 0.6285, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.1488095238095238, | |
| "eval_loss": 1.7950905561447144, | |
| "eval_runtime": 124.0984, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.15453296703296704, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 3.8615560640732266e-05, | |
| "loss": 0.6192, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.16025641025641027, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 4.0045766590389014e-05, | |
| "loss": 0.6121, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.16025641025641027, | |
| "eval_loss": 1.7816270589828491, | |
| "eval_runtime": 124.0951, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.16597985347985347, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 4.147597254004577e-05, | |
| "loss": 0.6039, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.1717032967032967, | |
| "grad_norm": 0.5625, | |
| "learning_rate": 4.290617848970252e-05, | |
| "loss": 0.5923, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1717032967032967, | |
| "eval_loss": 1.8131866455078125, | |
| "eval_runtime": 124.0957, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.17742673992673993, | |
| "grad_norm": 0.61328125, | |
| "learning_rate": 4.433638443935927e-05, | |
| "loss": 0.5973, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.18315018315018314, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 4.5766590389016025e-05, | |
| "loss": 0.5908, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.18315018315018314, | |
| "eval_loss": 1.7663753032684326, | |
| "eval_runtime": 124.103, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.18887362637362637, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 4.719679633867277e-05, | |
| "loss": 0.5869, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.1945970695970696, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 4.862700228832952e-05, | |
| "loss": 0.5662, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.1945970695970696, | |
| "eval_loss": 1.830661416053772, | |
| "eval_runtime": 124.107, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.20032051282051283, | |
| "grad_norm": 0.65234375, | |
| "learning_rate": 4.9999989894357565e-05, | |
| "loss": 0.5641, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.20604395604395603, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 4.999316889636665e-05, | |
| "loss": 0.5637, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.20604395604395603, | |
| "eval_loss": 1.7863534688949585, | |
| "eval_runtime": 124.0697, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.21176739926739926, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 4.9973719827852006e-05, | |
| "loss": 0.5566, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.2174908424908425, | |
| "grad_norm": 0.578125, | |
| "learning_rate": 4.994165251566713e-05, | |
| "loss": 0.5475, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.2174908424908425, | |
| "eval_loss": 1.7987805604934692, | |
| "eval_runtime": 124.1161, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.22321428571428573, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 4.98969831621704e-05, | |
| "loss": 0.5544, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.22893772893772893, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 4.98397343370387e-05, | |
| "loss": 0.5421, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22893772893772893, | |
| "eval_loss": 1.7876337766647339, | |
| "eval_runtime": 124.1072, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23466117216117216, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 4.976993496586383e-05, | |
| "loss": 0.5386, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.2403846153846154, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 4.968762031553753e-05, | |
| "loss": 0.529, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.2403846153846154, | |
| "eval_loss": 1.7660707235336304, | |
| "eval_runtime": 124.1155, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.24610805860805862, | |
| "grad_norm": 0.625, | |
| "learning_rate": 4.959283197643249e-05, | |
| "loss": 0.5278, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.2518315018315018, | |
| "grad_norm": 0.64453125, | |
| "learning_rate": 4.948561784138841e-05, | |
| "loss": 0.5202, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.2518315018315018, | |
| "eval_loss": 1.770936131477356, | |
| "eval_runtime": 124.1257, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.25755494505494503, | |
| "grad_norm": 0.59375, | |
| "learning_rate": 4.936603208151355e-05, | |
| "loss": 0.5071, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.2632783882783883, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 4.9234135118814246e-05, | |
| "loss": 0.5287, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.2632783882783883, | |
| "eval_loss": 1.7680959701538086, | |
| "eval_runtime": 124.1059, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.2690018315018315, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 4.908999359566602e-05, | |
| "loss": 0.5116, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.27472527472527475, | |
| "grad_norm": 0.61328125, | |
| "learning_rate": 4.8933680341141775e-05, | |
| "loss": 0.514, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.27472527472527475, | |
| "eval_loss": 1.7765259742736816, | |
| "eval_runtime": 124.1161, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.28044871794871795, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 4.8765274334214116e-05, | |
| "loss": 0.5099, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.28617216117216115, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 4.8584860663850404e-05, | |
| "loss": 0.5026, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.28617216117216115, | |
| "eval_loss": 1.7931022644042969, | |
| "eval_runtime": 124.0904, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2918956043956044, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 4.839253048602059e-05, | |
| "loss": 0.5044, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.2976190476190476, | |
| "grad_norm": 0.6484375, | |
| "learning_rate": 4.818838097763967e-05, | |
| "loss": 0.5038, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.2976190476190476, | |
| "eval_loss": 1.7807551622390747, | |
| "eval_runtime": 124.0961, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.3033424908424908, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 4.7972515287468e-05, | |
| "loss": 0.4828, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.3090659340659341, | |
| "grad_norm": 0.64453125, | |
| "learning_rate": 4.774504248399427e-05, | |
| "loss": 0.5052, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.3090659340659341, | |
| "eval_loss": 1.7688714265823364, | |
| "eval_runtime": 124.0947, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.3147893772893773, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 4.750607750032748e-05, | |
| "loss": 0.4938, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.32051282051282054, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 4.725574107612567e-05, | |
| "loss": 0.4918, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.32051282051282054, | |
| "eval_loss": 1.7862409353256226, | |
| "eval_runtime": 124.0923, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.32623626373626374, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 4.699415969659098e-05, | |
| "loss": 0.4847, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.33195970695970695, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 4.672146552856155e-05, | |
| "loss": 0.4817, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.33195970695970695, | |
| "eval_loss": 1.7916373014450073, | |
| "eval_runtime": 124.0857, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.3376831501831502, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 4.6437796353732824e-05, | |
| "loss": 0.4908, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.3434065934065934, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 4.614329549904187e-05, | |
| "loss": 0.4806, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3434065934065934, | |
| "eval_loss": 1.7795602083206177, | |
| "eval_runtime": 124.1973, | |
| "eval_samples_per_second": 2.254, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3491300366300366, | |
| "grad_norm": 0.65234375, | |
| "learning_rate": 4.583811176424984e-05, | |
| "loss": 0.4831, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.35485347985347987, | |
| "grad_norm": 0.6640625, | |
| "learning_rate": 4.5522399346759304e-05, | |
| "loss": 0.4849, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.35485347985347987, | |
| "eval_loss": 1.76537024974823, | |
| "eval_runtime": 124.1057, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.3605769230769231, | |
| "grad_norm": 0.6171875, | |
| "learning_rate": 4.51963177637043e-05, | |
| "loss": 0.4787, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.3663003663003663, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 4.4860031771352626e-05, | |
| "loss": 0.4784, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.3663003663003663, | |
| "eval_loss": 1.7576422691345215, | |
| "eval_runtime": 124.109, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.37202380952380953, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 4.451371128186087e-05, | |
| "loss": 0.485, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.37774725274725274, | |
| "grad_norm": 0.64453125, | |
| "learning_rate": 4.4157531277424503e-05, | |
| "loss": 0.4712, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.37774725274725274, | |
| "eval_loss": 1.7745938301086426, | |
| "eval_runtime": 124.0704, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.383470695970696, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 4.379167172186619e-05, | |
| "loss": 0.4899, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.3891941391941392, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 4.3416317469707125e-05, | |
| "loss": 0.4715, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.3891941391941392, | |
| "eval_loss": 1.7567591667175293, | |
| "eval_runtime": 124.0669, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.3949175824175824, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 4.3031658172767266e-05, | |
| "loss": 0.472, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.40064102564102566, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 4.263788818434168e-05, | |
| "loss": 0.4608, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.40064102564102566, | |
| "eval_loss": 1.7424110174179077, | |
| "eval_runtime": 124.063, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.40636446886446886, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 4.223520646100145e-05, | |
| "loss": 0.4586, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.41208791208791207, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 4.182381646206868e-05, | |
| "loss": 0.4629, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.41208791208791207, | |
| "eval_loss": 1.7561120986938477, | |
| "eval_runtime": 124.1336, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.4178113553113553, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 4.140392604681646e-05, | |
| "loss": 0.4655, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.42353479853479853, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 4.097574736944575e-05, | |
| "loss": 0.4591, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.42353479853479853, | |
| "eval_loss": 1.7497508525848389, | |
| "eval_runtime": 124.0856, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.42925824175824173, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 4.053949677189208e-05, | |
| "loss": 0.4658, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.434981684981685, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 4.0095394674516506e-05, | |
| "loss": 0.4652, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.434981684981685, | |
| "eval_loss": 1.736577033996582, | |
| "eval_runtime": 124.0959, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.4407051282051282, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 3.96436654647358e-05, | |
| "loss": 0.4588, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.44642857142857145, | |
| "grad_norm": 0.66796875, | |
| "learning_rate": 3.918453738364824e-05, | |
| "loss": 0.461, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.44642857142857145, | |
| "eval_loss": 1.7393635511398315, | |
| "eval_runtime": 124.0934, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.45215201465201466, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 3.871824241071236e-05, | |
| "loss": 0.4493, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.45787545787545786, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 3.824501614653676e-05, | |
| "loss": 0.4469, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.45787545787545786, | |
| "eval_loss": 1.7396734952926636, | |
| "eval_runtime": 124.0876, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4635989010989011, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 3.7765097693840385e-05, | |
| "loss": 0.4496, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.4693223443223443, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 3.727872953664322e-05, | |
| "loss": 0.4521, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.4693223443223443, | |
| "eval_loss": 1.7555357217788696, | |
| "eval_runtime": 124.0832, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.4750457875457875, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 3.678615741774861e-05, | |
| "loss": 0.4598, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 3.628763021457909e-05, | |
| "loss": 0.4498, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "eval_loss": 1.7651796340942383, | |
| "eval_runtime": 124.0848, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.486492673992674, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 3.57833998134283e-05, | |
| "loss": 0.4471, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.49221611721611724, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 3.5273720982192716e-05, | |
| "loss": 0.4541, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.49221611721611724, | |
| "eval_loss": 1.758300542831421, | |
| "eval_runtime": 124.0895, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.49793956043956045, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 3.475885124164737e-05, | |
| "loss": 0.4595, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.5036630036630036, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 3.4239050735330754e-05, | |
| "loss": 0.4594, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.5036630036630036, | |
| "eval_loss": 1.7604867219924927, | |
| "eval_runtime": 124.0898, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.5093864468864469, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 3.371458209810437e-05, | |
| "loss": 0.4584, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.5151098901098901, | |
| "grad_norm": 0.90625, | |
| "learning_rate": 3.3185710323453684e-05, | |
| "loss": 0.4514, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5151098901098901, | |
| "eval_loss": 1.7685655355453491, | |
| "eval_runtime": 124.0911, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5208333333333334, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 3.265270262959723e-05, | |
| "loss": 0.4523, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.5265567765567766, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 3.211582832447175e-05, | |
| "loss": 0.4395, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.5265567765567766, | |
| "eval_loss": 1.7713632583618164, | |
| "eval_runtime": 124.0855, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.5322802197802198, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 3.1575358669661356e-05, | |
| "loss": 0.4464, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.538003663003663, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 3.103156674333976e-05, | |
| "loss": 0.4384, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.538003663003663, | |
| "eval_loss": 1.788908839225769, | |
| "eval_runtime": 124.0945, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.5437271062271062, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 3.0484727302294475e-05, | |
| "loss": 0.4446, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.5494505494505495, | |
| "grad_norm": 0.6875, | |
| "learning_rate": 2.9935116643102983e-05, | |
| "loss": 0.4392, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5494505494505495, | |
| "eval_loss": 1.7709113359451294, | |
| "eval_runtime": 124.088, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5551739926739927, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 2.9383012462530895e-05, | |
| "loss": 0.4406, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.5608974358974359, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 2.8828693717222625e-05, | |
| "loss": 0.4495, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.5608974358974359, | |
| "eval_loss": 1.7554136514663696, | |
| "eval_runtime": 124.0986, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.5666208791208791, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 2.8272440482755535e-05, | |
| "loss": 0.4433, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 0.5723443223443223, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 2.771453381212865e-05, | |
| "loss": 0.4375, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5723443223443223, | |
| "eval_loss": 1.7531843185424805, | |
| "eval_runtime": 124.0918, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5780677655677655, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 2.715525559375764e-05, | |
| "loss": 0.4405, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 0.5837912087912088, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 2.6594888409047557e-05, | |
| "loss": 0.4441, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.5837912087912088, | |
| "eval_loss": 1.7770174741744995, | |
| "eval_runtime": 124.0878, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.589514652014652, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 2.6033715389615588e-05, | |
| "loss": 0.4403, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 2.5472020074235635e-05, | |
| "loss": 0.4458, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "eval_loss": 1.7528095245361328, | |
| "eval_runtime": 124.0953, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.6009615384615384, | |
| "grad_norm": 0.62890625, | |
| "learning_rate": 2.4910086265577364e-05, | |
| "loss": 0.4368, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.6066849816849816, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 2.4348197886811702e-05, | |
| "loss": 0.4343, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.6066849816849816, | |
| "eval_loss": 1.7645584344863892, | |
| "eval_runtime": 124.0831, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.612408424908425, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 2.3786638838155694e-05, | |
| "loss": 0.4444, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 0.6181318681318682, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 2.32256928534287e-05, | |
| "loss": 0.433, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.6181318681318682, | |
| "eval_loss": 1.7689203023910522, | |
| "eval_runtime": 124.0857, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.6238553113553114, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 2.2665643356692923e-05, | |
| "loss": 0.4344, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 0.6295787545787546, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 2.210677331905019e-05, | |
| "loss": 0.4371, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6295787545787546, | |
| "eval_loss": 1.7737501859664917, | |
| "eval_runtime": 124.0948, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6353021978021978, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 2.1549365115667853e-05, | |
| "loss": 0.4377, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 2.099370038310553e-05, | |
| "loss": 0.4376, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "eval_loss": 1.7633239030838013, | |
| "eval_runtime": 124.0669, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.6467490842490843, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 2.044005987701531e-05, | |
| "loss": 0.4388, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 0.6524725274725275, | |
| "grad_norm": 0.7734375, | |
| "learning_rate": 1.9888723330286763e-05, | |
| "loss": 0.4366, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.6524725274725275, | |
| "eval_loss": 1.7809503078460693, | |
| "eval_runtime": 124.1026, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.6581959706959707, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 1.933996931170898e-05, | |
| "loss": 0.4333, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.6639194139194139, | |
| "grad_norm": 0.7421875, | |
| "learning_rate": 1.879407508522056e-05, | |
| "loss": 0.43, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6639194139194139, | |
| "eval_loss": 1.7684820890426636, | |
| "eval_runtime": 124.0911, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6696428571428571, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 1.8251316469819075e-05, | |
| "loss": 0.4318, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 0.6753663003663004, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 1.7711967700200435e-05, | |
| "loss": 0.4345, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.6753663003663004, | |
| "eval_loss": 1.7761142253875732, | |
| "eval_runtime": 124.0752, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.6810897435897436, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 1.7176301288198894e-05, | |
| "loss": 0.4362, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 0.6868131868131868, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 1.6644587885097457e-05, | |
| "loss": 0.4379, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6868131868131868, | |
| "eval_loss": 1.7782317399978638, | |
| "eval_runtime": 124.0838, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.69253663003663, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 1.611709614487835e-05, | |
| "loss": 0.4379, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 0.6982600732600732, | |
| "grad_norm": 0.75, | |
| "learning_rate": 1.5594092588482718e-05, | |
| "loss": 0.4294, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.6982600732600732, | |
| "eval_loss": 1.7736785411834717, | |
| "eval_runtime": 124.0841, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.7039835164835165, | |
| "grad_norm": 0.75, | |
| "learning_rate": 1.5075841469147974e-05, | |
| "loss": 0.4356, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 0.7097069597069597, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 1.456260463889102e-05, | |
| "loss": 0.4441, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.7097069597069597, | |
| "eval_loss": 1.7646363973617554, | |
| "eval_runtime": 124.1036, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.7154304029304029, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 1.4054641416204609e-05, | |
| "loss": 0.4347, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.7211538461538461, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 1.3552208455033932e-05, | |
| "loss": 0.4396, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.7211538461538461, | |
| "eval_loss": 1.7779277563095093, | |
| "eval_runtime": 124.0892, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.7268772893772893, | |
| "grad_norm": 0.75, | |
| "learning_rate": 1.3055559615099353e-05, | |
| "loss": 0.4399, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 0.7326007326007326, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 1.256494583363104e-05, | |
| "loss": 0.4307, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.7326007326007326, | |
| "eval_loss": 1.7766470909118652, | |
| "eval_runtime": 124.1063, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.7383241758241759, | |
| "grad_norm": 0.79296875, | |
| "learning_rate": 1.2080614998580212e-05, | |
| "loss": 0.4351, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 0.7440476190476191, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 1.1602811823371069e-05, | |
| "loss": 0.4331, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7440476190476191, | |
| "eval_loss": 1.7733304500579834, | |
| "eval_runtime": 124.0732, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7497710622710623, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 1.1131777723256629e-05, | |
| "loss": 0.4373, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 0.7554945054945055, | |
| "grad_norm": 0.6484375, | |
| "learning_rate": 1.0667750693341072e-05, | |
| "loss": 0.4326, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.7554945054945055, | |
| "eval_loss": 1.7796032428741455, | |
| "eval_runtime": 124.0885, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.7612179487179487, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 1.0210965188330119e-05, | |
| "loss": 0.4314, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 0.766941391941392, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 9.761652004070241e-06, | |
| "loss": 0.4286, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.766941391941392, | |
| "eval_loss": 1.7802751064300537, | |
| "eval_runtime": 124.0901, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.7726648351648352, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 9.320038160936568e-06, | |
| "loss": 0.44, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.7783882783882784, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 8.886346789128305e-06, | |
| "loss": 0.4294, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7783882783882784, | |
| "eval_loss": 1.7787123918533325, | |
| "eval_runtime": 124.0869, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7841117216117216, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 8.460797015929873e-06, | |
| "loss": 0.4401, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 0.7898351648351648, | |
| "grad_norm": 0.71484375, | |
| "learning_rate": 8.04360385499437e-06, | |
| "loss": 0.4294, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.7898351648351648, | |
| "eval_loss": 1.779537558555603, | |
| "eval_runtime": 124.0971, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.795558608058608, | |
| "grad_norm": 0.80859375, | |
| "learning_rate": 7.63497809770566e-06, | |
| "loss": 0.4327, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 0.8012820512820513, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 7.235126206673651e-06, | |
| "loss": 0.4364, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8012820512820513, | |
| "eval_loss": 1.7765251398086548, | |
| "eval_runtime": 124.0761, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8070054945054945, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 6.844250211416903e-06, | |
| "loss": 0.4392, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 0.8127289377289377, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 6.462547606284947e-06, | |
| "loss": 0.4414, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.8127289377289377, | |
| "eval_loss": 1.7783187627792358, | |
| "eval_runtime": 124.1204, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.8184523809523809, | |
| "grad_norm": 0.7421875, | |
| "learning_rate": 6.09021125067217e-06, | |
| "loss": 0.4354, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 0.8241758241758241, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 5.7274292715734315e-06, | |
| "loss": 0.4336, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.8241758241758241, | |
| "eval_loss": 1.7745906114578247, | |
| "eval_runtime": 124.1664, | |
| "eval_samples_per_second": 2.255, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.8298992673992674, | |
| "grad_norm": 0.7421875, | |
| "learning_rate": 5.374384968530952e-06, | |
| "loss": 0.4296, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.8356227106227107, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 5.031256721020139e-06, | |
| "loss": 0.4324, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.8356227106227107, | |
| "eval_loss": 1.772754192352295, | |
| "eval_runtime": 124.0892, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.8413461538461539, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 4.698217898321483e-06, | |
| "loss": 0.4304, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 0.8470695970695971, | |
| "grad_norm": 0.71875, | |
| "learning_rate": 4.375436771923766e-06, | |
| "loss": 0.4414, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.8470695970695971, | |
| "eval_loss": 1.7765103578567505, | |
| "eval_runtime": 124.0978, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.8527930402930403, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 4.063076430503138e-06, | |
| "loss": 0.4258, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 0.8585164835164835, | |
| "grad_norm": 0.76953125, | |
| "learning_rate": 3.761294697520751e-06, | |
| "loss": 0.4288, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.8585164835164835, | |
| "eval_loss": 1.7792127132415771, | |
| "eval_runtime": 124.1222, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.8642399267399268, | |
| "grad_norm": 0.72265625, | |
| "learning_rate": 3.4702440514807366e-06, | |
| "loss": 0.4365, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 0.86996336996337, | |
| "grad_norm": 0.79296875, | |
| "learning_rate": 3.1900715488887873e-06, | |
| "loss": 0.4359, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.86996336996337, | |
| "eval_loss": 1.777583122253418, | |
| "eval_runtime": 124.1273, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.8756868131868132, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 2.9209187499502604e-06, | |
| "loss": 0.4388, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 0.8814102564102564, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 2.662921647045355e-06, | |
| "loss": 0.4242, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.8814102564102564, | |
| "eval_loss": 1.7761567831039429, | |
| "eval_runtime": 124.1269, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.8871336996336996, | |
| "grad_norm": 0.73828125, | |
| "learning_rate": 2.4162105960174486e-06, | |
| "loss": 0.4352, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 2.180910250309423e-06, | |
| "loss": 0.4413, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "eval_loss": 1.7751096487045288, | |
| "eval_runtime": 124.1103, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.8985805860805861, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 1.957139497981131e-06, | |
| "loss": 0.4293, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 0.9043040293040293, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 1.74501140163994e-06, | |
| "loss": 0.4402, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.9043040293040293, | |
| "eval_loss": 1.7753708362579346, | |
| "eval_runtime": 124.1127, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.9100274725274725, | |
| "grad_norm": 0.75, | |
| "learning_rate": 1.5446331413145887e-06, | |
| "loss": 0.4436, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 0.9157509157509157, | |
| "grad_norm": 0.703125, | |
| "learning_rate": 1.3561059603013265e-06, | |
| "loss": 0.4452, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9157509157509157, | |
| "eval_loss": 1.7749762535095215, | |
| "eval_runtime": 124.0977, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9214743589743589, | |
| "grad_norm": 0.6640625, | |
| "learning_rate": 1.1795251140096358e-06, | |
| "loss": 0.4458, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 0.9271978021978022, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 1.014979821833395e-06, | |
| "loss": 0.4346, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.9271978021978022, | |
| "eval_loss": 1.775481104850769, | |
| "eval_runtime": 124.1005, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.9329212454212454, | |
| "grad_norm": 0.73046875, | |
| "learning_rate": 8.625532220718186e-07, | |
| "loss": 0.4361, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 0.9386446886446886, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 7.223223299229198e-07, | |
| "loss": 0.4396, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.9386446886446886, | |
| "eval_loss": 1.7751343250274658, | |
| "eval_runtime": 124.089, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.9443681318681318, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 5.943579985707409e-07, | |
| "loss": 0.4337, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.950091575091575, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 4.787248833860114e-07, | |
| "loss": 0.44, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.950091575091575, | |
| "eval_loss": 1.7752093076705933, | |
| "eval_runtime": 124.0951, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.9558150183150184, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 3.7548140925833806e-07, | |
| "loss": 0.4362, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 2.8467974107636017e-07, | |
| "loss": 0.4333, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "eval_loss": 1.7753241062164307, | |
| "eval_runtime": 124.0863, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9672619047619048, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 2.063657573708966e-07, | |
| "loss": 0.4338, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 0.972985347985348, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 1.4057902713431327e-07, | |
| "loss": 0.4348, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.972985347985348, | |
| "eval_loss": 1.7753559350967407, | |
| "eval_runtime": 124.0677, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9787087912087912, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 8.735278982785755e-08, | |
| "loss": 0.4306, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 0.9844322344322345, | |
| "grad_norm": 0.74609375, | |
| "learning_rate": 4.671393858705908e-08, | |
| "loss": 0.4331, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.9844322344322345, | |
| "eval_loss": 1.77518630027771, | |
| "eval_runtime": 124.0873, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.9901556776556777, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 1.868300663367406e-08, | |
| "loss": 0.4329, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 0.9958791208791209, | |
| "grad_norm": 0.76171875, | |
| "learning_rate": 3.2741569010674712e-09, | |
| "loss": 0.4326, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.9958791208791209, | |
| "eval_loss": 1.7753793001174927, | |
| "eval_runtime": 124.103, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.564, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 8736, | |
| "total_flos": 1.6067287852253184e+18, | |
| "train_loss": 0.5878726873860691, | |
| "train_runtime": 61938.6999, | |
| "train_samples_per_second": 0.564, | |
| "train_steps_per_second": 0.141 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 8736, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 250, | |
| "total_flos": 1.6067287852253184e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |