| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1695906432748537, | |
| "eval_steps": 100, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005847953216374269, | |
| "grad_norm": 19.25, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.3094, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.011695906432748537, | |
| "grad_norm": 20.515625, | |
| "learning_rate": 2.4874371859296484e-05, | |
| "loss": 0.6781, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.017543859649122806, | |
| "grad_norm": 9.015625, | |
| "learning_rate": 2.4748743718592964e-05, | |
| "loss": 0.6127, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.023391812865497075, | |
| "grad_norm": 9.4765625, | |
| "learning_rate": 2.462311557788945e-05, | |
| "loss": 0.6338, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.023391812865497075, | |
| "eval_loss": 0.6474161744117737, | |
| "eval_runtime": 125.9492, | |
| "eval_samples_per_second": 3.573, | |
| "eval_steps_per_second": 0.453, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.029239766081871343, | |
| "grad_norm": 16.71875, | |
| "learning_rate": 2.449748743718593e-05, | |
| "loss": 0.506, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03508771929824561, | |
| "grad_norm": 21.125, | |
| "learning_rate": 2.4371859296482413e-05, | |
| "loss": 0.501, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04093567251461988, | |
| "grad_norm": 17.953125, | |
| "learning_rate": 2.4246231155778896e-05, | |
| "loss": 0.6208, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04678362573099415, | |
| "grad_norm": 7.9609375, | |
| "learning_rate": 2.4120603015075376e-05, | |
| "loss": 0.5748, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04678362573099415, | |
| "eval_loss": 0.5898649096488953, | |
| "eval_runtime": 129.2682, | |
| "eval_samples_per_second": 3.481, | |
| "eval_steps_per_second": 0.441, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05263157894736842, | |
| "grad_norm": 7.4921875, | |
| "learning_rate": 2.3994974874371863e-05, | |
| "loss": 0.5821, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05847953216374269, | |
| "grad_norm": 8.703125, | |
| "learning_rate": 2.3869346733668342e-05, | |
| "loss": 0.5764, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06432748538011696, | |
| "grad_norm": 10.3828125, | |
| "learning_rate": 2.3743718592964825e-05, | |
| "loss": 0.5921, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 14.90625, | |
| "learning_rate": 2.361809045226131e-05, | |
| "loss": 0.4678, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "eval_loss": 0.5602560639381409, | |
| "eval_runtime": 129.0871, | |
| "eval_samples_per_second": 3.486, | |
| "eval_steps_per_second": 0.442, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07602339181286549, | |
| "grad_norm": 16.6875, | |
| "learning_rate": 2.3492462311557788e-05, | |
| "loss": 0.5118, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.08187134502923976, | |
| "grad_norm": 17.921875, | |
| "learning_rate": 2.3366834170854275e-05, | |
| "loss": 0.5199, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.08771929824561403, | |
| "grad_norm": 16.5625, | |
| "learning_rate": 2.3241206030150754e-05, | |
| "loss": 0.4611, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.0935672514619883, | |
| "grad_norm": 10.5, | |
| "learning_rate": 2.3115577889447238e-05, | |
| "loss": 0.428, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.0935672514619883, | |
| "eval_loss": 0.524933397769928, | |
| "eval_runtime": 127.1828, | |
| "eval_samples_per_second": 3.538, | |
| "eval_steps_per_second": 0.448, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09941520467836257, | |
| "grad_norm": 11.6328125, | |
| "learning_rate": 2.298994974874372e-05, | |
| "loss": 0.4953, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.10526315789473684, | |
| "grad_norm": 5.4296875, | |
| "learning_rate": 2.28643216080402e-05, | |
| "loss": 0.4173, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 2.2738693467336687e-05, | |
| "loss": 0.4225, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.11695906432748537, | |
| "grad_norm": 8.84375, | |
| "learning_rate": 2.2613065326633167e-05, | |
| "loss": 0.3798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11695906432748537, | |
| "eval_loss": 0.49835124611854553, | |
| "eval_runtime": 128.3365, | |
| "eval_samples_per_second": 3.506, | |
| "eval_steps_per_second": 0.444, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12280701754385964, | |
| "grad_norm": 11.4609375, | |
| "learning_rate": 2.248743718592965e-05, | |
| "loss": 0.4386, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.1286549707602339, | |
| "grad_norm": 16.4375, | |
| "learning_rate": 2.2361809045226133e-05, | |
| "loss": 0.4537, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.13450292397660818, | |
| "grad_norm": 10.140625, | |
| "learning_rate": 2.2236180904522613e-05, | |
| "loss": 0.3514, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 3.265625, | |
| "learning_rate": 2.21105527638191e-05, | |
| "loss": 0.3665, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "eval_loss": 0.47331658005714417, | |
| "eval_runtime": 128.8126, | |
| "eval_samples_per_second": 3.493, | |
| "eval_steps_per_second": 0.443, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.14619883040935672, | |
| "grad_norm": 12.3046875, | |
| "learning_rate": 2.198492462311558e-05, | |
| "loss": 0.2409, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.15204678362573099, | |
| "grad_norm": 9.703125, | |
| "learning_rate": 2.1859296482412062e-05, | |
| "loss": 0.4553, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 10.65625, | |
| "learning_rate": 2.1733668341708545e-05, | |
| "loss": 0.4242, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.16374269005847952, | |
| "grad_norm": 4.96484375, | |
| "learning_rate": 2.1608040201005025e-05, | |
| "loss": 0.4406, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.16374269005847952, | |
| "eval_loss": 0.45101454854011536, | |
| "eval_runtime": 129.1251, | |
| "eval_samples_per_second": 3.485, | |
| "eval_steps_per_second": 0.441, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1695906432748538, | |
| "grad_norm": 6.20703125, | |
| "learning_rate": 2.1482412060301508e-05, | |
| "loss": 0.408, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.17543859649122806, | |
| "grad_norm": 9.875, | |
| "learning_rate": 2.135678391959799e-05, | |
| "loss": 0.5067, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.18128654970760233, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 2.1231155778894474e-05, | |
| "loss": 0.4338, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.1871345029239766, | |
| "grad_norm": 12.765625, | |
| "learning_rate": 2.1105527638190957e-05, | |
| "loss": 0.4723, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.1871345029239766, | |
| "eval_loss": 0.4244983494281769, | |
| "eval_runtime": 127.3581, | |
| "eval_samples_per_second": 3.533, | |
| "eval_steps_per_second": 0.448, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.19298245614035087, | |
| "grad_norm": 26.03125, | |
| "learning_rate": 2.0979899497487437e-05, | |
| "loss": 0.4424, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.19883040935672514, | |
| "grad_norm": 5.234375, | |
| "learning_rate": 2.085427135678392e-05, | |
| "loss": 0.3599, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.2046783625730994, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 2.0728643216080403e-05, | |
| "loss": 0.3399, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 9.6640625, | |
| "learning_rate": 2.0603015075376886e-05, | |
| "loss": 0.4807, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "eval_loss": 0.41580215096473694, | |
| "eval_runtime": 128.5483, | |
| "eval_samples_per_second": 3.501, | |
| "eval_steps_per_second": 0.443, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.21637426900584794, | |
| "grad_norm": 9.03125, | |
| "learning_rate": 2.047738693467337e-05, | |
| "loss": 0.3341, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 10.3515625, | |
| "learning_rate": 2.035175879396985e-05, | |
| "loss": 0.4407, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.22807017543859648, | |
| "grad_norm": 13.046875, | |
| "learning_rate": 2.0226130653266332e-05, | |
| "loss": 0.391, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.23391812865497075, | |
| "grad_norm": 13.0859375, | |
| "learning_rate": 2.0100502512562815e-05, | |
| "loss": 0.4196, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23391812865497075, | |
| "eval_loss": 0.39709773659706116, | |
| "eval_runtime": 129.6284, | |
| "eval_samples_per_second": 3.471, | |
| "eval_steps_per_second": 0.44, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23976608187134502, | |
| "grad_norm": 2.962890625, | |
| "learning_rate": 1.9974874371859298e-05, | |
| "loss": 0.3042, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.24561403508771928, | |
| "grad_norm": 11.9765625, | |
| "learning_rate": 1.984924623115578e-05, | |
| "loss": 0.3731, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.25146198830409355, | |
| "grad_norm": 5.42578125, | |
| "learning_rate": 1.972361809045226e-05, | |
| "loss": 0.3032, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.2573099415204678, | |
| "grad_norm": 8.9140625, | |
| "learning_rate": 1.9597989949748744e-05, | |
| "loss": 0.3443, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.2573099415204678, | |
| "eval_loss": 0.3737768530845642, | |
| "eval_runtime": 126.3694, | |
| "eval_samples_per_second": 3.561, | |
| "eval_steps_per_second": 0.451, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "grad_norm": 14.0625, | |
| "learning_rate": 1.9472361809045227e-05, | |
| "loss": 0.2804, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.26900584795321636, | |
| "grad_norm": 5.19921875, | |
| "learning_rate": 1.934673366834171e-05, | |
| "loss": 0.3257, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.27485380116959063, | |
| "grad_norm": 9.78125, | |
| "learning_rate": 1.9221105527638193e-05, | |
| "loss": 0.3625, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 12.890625, | |
| "learning_rate": 1.9095477386934673e-05, | |
| "loss": 0.4133, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "eval_loss": 0.3631095290184021, | |
| "eval_runtime": 125.638, | |
| "eval_samples_per_second": 3.582, | |
| "eval_steps_per_second": 0.454, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.28654970760233917, | |
| "grad_norm": 7.71875, | |
| "learning_rate": 1.8969849246231156e-05, | |
| "loss": 0.5163, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.29239766081871343, | |
| "grad_norm": 7.25390625, | |
| "learning_rate": 1.884422110552764e-05, | |
| "loss": 0.2991, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.2982456140350877, | |
| "grad_norm": 6.5234375, | |
| "learning_rate": 1.8718592964824123e-05, | |
| "loss": 0.431, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.30409356725146197, | |
| "grad_norm": 3.8515625, | |
| "learning_rate": 1.8592964824120602e-05, | |
| "loss": 0.2838, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.30409356725146197, | |
| "eval_loss": 0.33338162302970886, | |
| "eval_runtime": 127.3115, | |
| "eval_samples_per_second": 3.535, | |
| "eval_steps_per_second": 0.448, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.30994152046783624, | |
| "grad_norm": 9.828125, | |
| "learning_rate": 1.8467336683417085e-05, | |
| "loss": 0.2653, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 5.34765625, | |
| "learning_rate": 1.834170854271357e-05, | |
| "loss": 0.2515, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.3216374269005848, | |
| "grad_norm": 12.15625, | |
| "learning_rate": 1.821608040201005e-05, | |
| "loss": 0.2983, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.32748538011695905, | |
| "grad_norm": 9.171875, | |
| "learning_rate": 1.8090452261306535e-05, | |
| "loss": 0.4134, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.32748538011695905, | |
| "eval_loss": 0.32642024755477905, | |
| "eval_runtime": 127.196, | |
| "eval_samples_per_second": 3.538, | |
| "eval_steps_per_second": 0.448, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 7.16015625, | |
| "learning_rate": 1.7964824120603014e-05, | |
| "loss": 0.3015, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.3391812865497076, | |
| "grad_norm": 15.6640625, | |
| "learning_rate": 1.7839195979899497e-05, | |
| "loss": 0.3098, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.34502923976608185, | |
| "grad_norm": 13.8984375, | |
| "learning_rate": 1.771356783919598e-05, | |
| "loss": 0.4022, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 9.90625, | |
| "learning_rate": 1.7587939698492464e-05, | |
| "loss": 0.2838, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "eval_loss": 0.3124904930591583, | |
| "eval_runtime": 128.6196, | |
| "eval_samples_per_second": 3.499, | |
| "eval_steps_per_second": 0.443, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3567251461988304, | |
| "grad_norm": 18.234375, | |
| "learning_rate": 1.7462311557788947e-05, | |
| "loss": 0.3792, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.36257309941520466, | |
| "grad_norm": 1.482421875, | |
| "learning_rate": 1.7336683417085427e-05, | |
| "loss": 0.4708, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.3684210526315789, | |
| "grad_norm": 10.8046875, | |
| "learning_rate": 1.721105527638191e-05, | |
| "loss": 0.3695, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.3742690058479532, | |
| "grad_norm": 13.6328125, | |
| "learning_rate": 1.7085427135678393e-05, | |
| "loss": 0.275, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.3742690058479532, | |
| "eval_loss": 0.294376403093338, | |
| "eval_runtime": 125.5965, | |
| "eval_samples_per_second": 3.583, | |
| "eval_steps_per_second": 0.454, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.38011695906432746, | |
| "grad_norm": 10.3359375, | |
| "learning_rate": 1.6959798994974876e-05, | |
| "loss": 0.3134, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.38596491228070173, | |
| "grad_norm": 4.671875, | |
| "learning_rate": 1.683417085427136e-05, | |
| "loss": 0.1961, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.391812865497076, | |
| "grad_norm": 3.734375, | |
| "learning_rate": 1.670854271356784e-05, | |
| "loss": 0.2686, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.39766081871345027, | |
| "grad_norm": 11.5859375, | |
| "learning_rate": 1.6582914572864322e-05, | |
| "loss": 0.4141, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.39766081871345027, | |
| "eval_loss": 0.2839096784591675, | |
| "eval_runtime": 126.4222, | |
| "eval_samples_per_second": 3.56, | |
| "eval_steps_per_second": 0.451, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.40350877192982454, | |
| "grad_norm": 12.0234375, | |
| "learning_rate": 1.6457286432160805e-05, | |
| "loss": 0.2948, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.4093567251461988, | |
| "grad_norm": 6.44140625, | |
| "learning_rate": 1.6331658291457288e-05, | |
| "loss": 0.3546, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.4152046783625731, | |
| "grad_norm": 7.5234375, | |
| "learning_rate": 1.620603015075377e-05, | |
| "loss": 0.3239, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 4.30078125, | |
| "learning_rate": 1.608040201005025e-05, | |
| "loss": 0.2498, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "eval_loss": 0.2749183773994446, | |
| "eval_runtime": 126.776, | |
| "eval_samples_per_second": 3.55, | |
| "eval_steps_per_second": 0.45, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.4269005847953216, | |
| "grad_norm": 8.4375, | |
| "learning_rate": 1.5954773869346734e-05, | |
| "loss": 0.3026, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.4327485380116959, | |
| "grad_norm": 8.140625, | |
| "learning_rate": 1.5829145728643217e-05, | |
| "loss": 0.3848, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.43859649122807015, | |
| "grad_norm": 10.046875, | |
| "learning_rate": 1.57035175879397e-05, | |
| "loss": 0.2574, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 9.1640625, | |
| "learning_rate": 1.5577889447236183e-05, | |
| "loss": 0.2817, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "eval_loss": 0.2605937719345093, | |
| "eval_runtime": 129.5889, | |
| "eval_samples_per_second": 3.473, | |
| "eval_steps_per_second": 0.44, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.4502923976608187, | |
| "grad_norm": 11.3828125, | |
| "learning_rate": 1.5452261306532663e-05, | |
| "loss": 0.376, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.45614035087719296, | |
| "grad_norm": 9.6953125, | |
| "learning_rate": 1.5326633165829146e-05, | |
| "loss": 0.2708, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.4619883040935672, | |
| "grad_norm": 10.0859375, | |
| "learning_rate": 1.5201005025125627e-05, | |
| "loss": 0.2509, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.4678362573099415, | |
| "grad_norm": 11.4921875, | |
| "learning_rate": 1.507537688442211e-05, | |
| "loss": 0.2899, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4678362573099415, | |
| "eval_loss": 0.2526080906391144, | |
| "eval_runtime": 127.4275, | |
| "eval_samples_per_second": 3.531, | |
| "eval_steps_per_second": 0.447, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 10.1796875, | |
| "learning_rate": 1.4949748743718595e-05, | |
| "loss": 0.3491, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.47953216374269003, | |
| "grad_norm": 4.72265625, | |
| "learning_rate": 1.4824120603015077e-05, | |
| "loss": 0.1925, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.4853801169590643, | |
| "grad_norm": 6.58203125, | |
| "learning_rate": 1.4698492462311558e-05, | |
| "loss": 0.2187, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "grad_norm": 11.4609375, | |
| "learning_rate": 1.457286432160804e-05, | |
| "loss": 0.2695, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "eval_loss": 0.2520677149295807, | |
| "eval_runtime": 129.417, | |
| "eval_samples_per_second": 3.477, | |
| "eval_steps_per_second": 0.44, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.49707602339181284, | |
| "grad_norm": 4.5, | |
| "learning_rate": 1.4447236180904523e-05, | |
| "loss": 0.1898, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.5029239766081871, | |
| "grad_norm": 7.640625, | |
| "learning_rate": 1.4321608040201007e-05, | |
| "loss": 0.3, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.5087719298245614, | |
| "grad_norm": 3.82421875, | |
| "learning_rate": 1.4195979899497489e-05, | |
| "loss": 0.2006, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 0.5146198830409356, | |
| "grad_norm": 3.509765625, | |
| "learning_rate": 1.407035175879397e-05, | |
| "loss": 0.2619, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.5146198830409356, | |
| "eval_loss": 0.24239127337932587, | |
| "eval_runtime": 125.4976, | |
| "eval_samples_per_second": 3.586, | |
| "eval_steps_per_second": 0.454, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.52046783625731, | |
| "grad_norm": 7.35546875, | |
| "learning_rate": 1.3944723618090452e-05, | |
| "loss": 0.3202, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 5.37109375, | |
| "learning_rate": 1.3819095477386935e-05, | |
| "loss": 0.143, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.5321637426900585, | |
| "grad_norm": 13.203125, | |
| "learning_rate": 1.369346733668342e-05, | |
| "loss": 0.3726, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 0.5380116959064327, | |
| "grad_norm": 7.08984375, | |
| "learning_rate": 1.3567839195979901e-05, | |
| "loss": 0.2238, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.5380116959064327, | |
| "eval_loss": 0.2372661828994751, | |
| "eval_runtime": 125.8271, | |
| "eval_samples_per_second": 3.576, | |
| "eval_steps_per_second": 0.453, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.543859649122807, | |
| "grad_norm": 7.73046875, | |
| "learning_rate": 1.3442211055276382e-05, | |
| "loss": 0.2301, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 0.5497076023391813, | |
| "grad_norm": 1.9326171875, | |
| "learning_rate": 1.3316582914572864e-05, | |
| "loss": 0.2891, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 5.25, | |
| "learning_rate": 1.3190954773869347e-05, | |
| "loss": 0.1861, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 15.7265625, | |
| "learning_rate": 1.306532663316583e-05, | |
| "loss": 0.3049, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "eval_loss": 0.23007704317569733, | |
| "eval_runtime": 125.6345, | |
| "eval_samples_per_second": 3.582, | |
| "eval_steps_per_second": 0.454, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.5672514619883041, | |
| "grad_norm": 12.0078125, | |
| "learning_rate": 1.2939698492462313e-05, | |
| "loss": 0.2584, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 0.5730994152046783, | |
| "grad_norm": 11.8125, | |
| "learning_rate": 1.2814070351758795e-05, | |
| "loss": 0.29, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.5789473684210527, | |
| "grad_norm": 4.7578125, | |
| "learning_rate": 1.2688442211055276e-05, | |
| "loss": 0.2648, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 0.5847953216374269, | |
| "grad_norm": 10.53125, | |
| "learning_rate": 1.2562814070351759e-05, | |
| "loss": 0.1308, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5847953216374269, | |
| "eval_loss": 0.2292058765888214, | |
| "eval_runtime": 127.785, | |
| "eval_samples_per_second": 3.522, | |
| "eval_steps_per_second": 0.446, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5906432748538012, | |
| "grad_norm": 8.421875, | |
| "learning_rate": 1.2437185929648242e-05, | |
| "loss": 0.3249, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 0.5964912280701754, | |
| "grad_norm": 10.8828125, | |
| "learning_rate": 1.2311557788944725e-05, | |
| "loss": 0.2164, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.6023391812865497, | |
| "grad_norm": 7.421875, | |
| "learning_rate": 1.2185929648241207e-05, | |
| "loss": 0.2512, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 0.6081871345029239, | |
| "grad_norm": 9.6171875, | |
| "learning_rate": 1.2060301507537688e-05, | |
| "loss": 0.1936, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.6081871345029239, | |
| "eval_loss": 0.21902820467948914, | |
| "eval_runtime": 125.3191, | |
| "eval_samples_per_second": 3.591, | |
| "eval_steps_per_second": 0.455, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.6140350877192983, | |
| "grad_norm": 1.091796875, | |
| "learning_rate": 1.1934673366834171e-05, | |
| "loss": 0.189, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 0.6198830409356725, | |
| "grad_norm": 7.83203125, | |
| "learning_rate": 1.1809045226130654e-05, | |
| "loss": 0.2179, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.6257309941520468, | |
| "grad_norm": 8.2109375, | |
| "learning_rate": 1.1683417085427137e-05, | |
| "loss": 0.224, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 5.1640625, | |
| "learning_rate": 1.1557788944723619e-05, | |
| "loss": 0.2479, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "eval_loss": 0.21907079219818115, | |
| "eval_runtime": 126.3038, | |
| "eval_samples_per_second": 3.563, | |
| "eval_steps_per_second": 0.451, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.6374269005847953, | |
| "grad_norm": 4.47265625, | |
| "learning_rate": 1.14321608040201e-05, | |
| "loss": 0.219, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 0.6432748538011696, | |
| "grad_norm": 3.30078125, | |
| "learning_rate": 1.1306532663316583e-05, | |
| "loss": 0.2741, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.6491228070175439, | |
| "grad_norm": 4.21484375, | |
| "learning_rate": 1.1180904522613066e-05, | |
| "loss": 0.1153, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 0.6549707602339181, | |
| "grad_norm": 5.125, | |
| "learning_rate": 1.105527638190955e-05, | |
| "loss": 0.1575, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.6549707602339181, | |
| "eval_loss": 0.2165260761976242, | |
| "eval_runtime": 125.8096, | |
| "eval_samples_per_second": 3.577, | |
| "eval_steps_per_second": 0.453, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.6608187134502924, | |
| "grad_norm": 0.83837890625, | |
| "learning_rate": 1.0929648241206031e-05, | |
| "loss": 0.2238, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 6.61328125, | |
| "learning_rate": 1.0804020100502512e-05, | |
| "loss": 0.3554, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.672514619883041, | |
| "grad_norm": 12.875, | |
| "learning_rate": 1.0678391959798995e-05, | |
| "loss": 0.096, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 0.6783625730994152, | |
| "grad_norm": 3.28125, | |
| "learning_rate": 1.0552763819095479e-05, | |
| "loss": 0.193, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.6783625730994152, | |
| "eval_loss": 0.21065188944339752, | |
| "eval_runtime": 125.6087, | |
| "eval_samples_per_second": 3.583, | |
| "eval_steps_per_second": 0.454, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.6842105263157895, | |
| "grad_norm": 8.7265625, | |
| "learning_rate": 1.042713567839196e-05, | |
| "loss": 0.1936, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 0.6900584795321637, | |
| "grad_norm": 4.0859375, | |
| "learning_rate": 1.0301507537688443e-05, | |
| "loss": 0.2103, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.695906432748538, | |
| "grad_norm": 15.5703125, | |
| "learning_rate": 1.0175879396984924e-05, | |
| "loss": 0.2701, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 3.73046875, | |
| "learning_rate": 1.0050251256281408e-05, | |
| "loss": 0.2526, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "eval_loss": 0.21144379675388336, | |
| "eval_runtime": 128.1153, | |
| "eval_samples_per_second": 3.512, | |
| "eval_steps_per_second": 0.445, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7076023391812866, | |
| "grad_norm": 1.537109375, | |
| "learning_rate": 9.92462311557789e-06, | |
| "loss": 0.2087, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 0.7134502923976608, | |
| "grad_norm": 0.55908203125, | |
| "learning_rate": 9.798994974874372e-06, | |
| "loss": 0.1532, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.7192982456140351, | |
| "grad_norm": 9.7578125, | |
| "learning_rate": 9.673366834170855e-06, | |
| "loss": 0.1985, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 0.7251461988304093, | |
| "grad_norm": 9.390625, | |
| "learning_rate": 9.547738693467337e-06, | |
| "loss": 0.1574, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.7251461988304093, | |
| "eval_loss": 0.20868618786334991, | |
| "eval_runtime": 129.4237, | |
| "eval_samples_per_second": 3.477, | |
| "eval_steps_per_second": 0.44, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.7309941520467836, | |
| "grad_norm": 4.6171875, | |
| "learning_rate": 9.42211055276382e-06, | |
| "loss": 0.2921, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 0.7368421052631579, | |
| "grad_norm": 10.859375, | |
| "learning_rate": 9.296482412060301e-06, | |
| "loss": 0.1932, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.7426900584795322, | |
| "grad_norm": 5.97265625, | |
| "learning_rate": 9.170854271356784e-06, | |
| "loss": 0.2999, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 0.7485380116959064, | |
| "grad_norm": 4.85546875, | |
| "learning_rate": 9.045226130653267e-06, | |
| "loss": 0.1989, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.7485380116959064, | |
| "eval_loss": 0.20511174201965332, | |
| "eval_runtime": 127.5209, | |
| "eval_samples_per_second": 3.529, | |
| "eval_steps_per_second": 0.447, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.7543859649122807, | |
| "grad_norm": 8.7265625, | |
| "learning_rate": 8.919597989949749e-06, | |
| "loss": 0.1795, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 0.7602339181286549, | |
| "grad_norm": 3.490234375, | |
| "learning_rate": 8.793969849246232e-06, | |
| "loss": 0.2444, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.7660818713450293, | |
| "grad_norm": 6.671875, | |
| "learning_rate": 8.668341708542713e-06, | |
| "loss": 0.2775, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 0.7719298245614035, | |
| "grad_norm": 5.1796875, | |
| "learning_rate": 8.542713567839196e-06, | |
| "loss": 0.1761, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.7719298245614035, | |
| "eval_loss": 0.20133711397647858, | |
| "eval_runtime": 129.2927, | |
| "eval_samples_per_second": 3.48, | |
| "eval_steps_per_second": 0.441, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 0.83935546875, | |
| "learning_rate": 8.41708542713568e-06, | |
| "loss": 0.216, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 0.783625730994152, | |
| "grad_norm": 2.580078125, | |
| "learning_rate": 8.291457286432161e-06, | |
| "loss": 0.0873, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 5.87109375, | |
| "learning_rate": 8.165829145728644e-06, | |
| "loss": 0.1145, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 0.7953216374269005, | |
| "grad_norm": 7.125, | |
| "learning_rate": 8.040201005025125e-06, | |
| "loss": 0.2223, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.7953216374269005, | |
| "eval_loss": 0.19959186017513275, | |
| "eval_runtime": 125.4382, | |
| "eval_samples_per_second": 3.587, | |
| "eval_steps_per_second": 0.454, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.8011695906432749, | |
| "grad_norm": 3.94921875, | |
| "learning_rate": 7.914572864321608e-06, | |
| "loss": 0.1845, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 0.8070175438596491, | |
| "grad_norm": 5.99609375, | |
| "learning_rate": 7.788944723618092e-06, | |
| "loss": 0.104, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.8128654970760234, | |
| "grad_norm": 7.953125, | |
| "learning_rate": 7.663316582914573e-06, | |
| "loss": 0.1119, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 0.8187134502923976, | |
| "grad_norm": 7.5859375, | |
| "learning_rate": 7.537688442211055e-06, | |
| "loss": 0.2127, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8187134502923976, | |
| "eval_loss": 0.1966124027967453, | |
| "eval_runtime": 126.749, | |
| "eval_samples_per_second": 3.55, | |
| "eval_steps_per_second": 0.45, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8245614035087719, | |
| "grad_norm": 25.390625, | |
| "learning_rate": 7.412060301507538e-06, | |
| "loss": 0.1597, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 0.8304093567251462, | |
| "grad_norm": 7.55078125, | |
| "learning_rate": 7.28643216080402e-06, | |
| "loss": 0.0942, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.8362573099415205, | |
| "grad_norm": 6.5390625, | |
| "learning_rate": 7.160804020100504e-06, | |
| "loss": 0.236, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 9.390625, | |
| "learning_rate": 7.035175879396985e-06, | |
| "loss": 0.2477, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "eval_loss": 0.1922728419303894, | |
| "eval_runtime": 125.5954, | |
| "eval_samples_per_second": 3.583, | |
| "eval_steps_per_second": 0.454, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.847953216374269, | |
| "grad_norm": 0.7783203125, | |
| "learning_rate": 6.909547738693467e-06, | |
| "loss": 0.3097, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 0.8538011695906432, | |
| "grad_norm": 5.91796875, | |
| "learning_rate": 6.7839195979899505e-06, | |
| "loss": 0.2097, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.8596491228070176, | |
| "grad_norm": 2.55078125, | |
| "learning_rate": 6.658291457286432e-06, | |
| "loss": 0.1837, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 0.8654970760233918, | |
| "grad_norm": 5.1015625, | |
| "learning_rate": 6.532663316582915e-06, | |
| "loss": 0.1931, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.8654970760233918, | |
| "eval_loss": 0.1908126324415207, | |
| "eval_runtime": 125.264, | |
| "eval_samples_per_second": 3.592, | |
| "eval_steps_per_second": 0.455, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.8713450292397661, | |
| "grad_norm": 1.0830078125, | |
| "learning_rate": 6.407035175879397e-06, | |
| "loss": 0.1688, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 0.8771929824561403, | |
| "grad_norm": 2.021484375, | |
| "learning_rate": 6.2814070351758795e-06, | |
| "loss": 0.1635, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.8830409356725146, | |
| "grad_norm": 24.90625, | |
| "learning_rate": 6.155778894472363e-06, | |
| "loss": 0.2588, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 4.9453125, | |
| "learning_rate": 6.030150753768844e-06, | |
| "loss": 0.182, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "eval_loss": 0.18878790736198425, | |
| "eval_runtime": 126.7176, | |
| "eval_samples_per_second": 3.551, | |
| "eval_steps_per_second": 0.45, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.8947368421052632, | |
| "grad_norm": 7.25, | |
| "learning_rate": 5.904522613065327e-06, | |
| "loss": 0.2793, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 0.9005847953216374, | |
| "grad_norm": 9.59375, | |
| "learning_rate": 5.778894472361809e-06, | |
| "loss": 0.1608, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.9064327485380117, | |
| "grad_norm": 2.943359375, | |
| "learning_rate": 5.653266331658292e-06, | |
| "loss": 0.2136, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 0.9122807017543859, | |
| "grad_norm": 0.48681640625, | |
| "learning_rate": 5.527638190954775e-06, | |
| "loss": 0.1693, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.9122807017543859, | |
| "eval_loss": 0.18779730796813965, | |
| "eval_runtime": 126.6799, | |
| "eval_samples_per_second": 3.552, | |
| "eval_steps_per_second": 0.45, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.9181286549707602, | |
| "grad_norm": 3.2109375, | |
| "learning_rate": 5.402010050251256e-06, | |
| "loss": 0.0918, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 0.9239766081871345, | |
| "grad_norm": 1.541015625, | |
| "learning_rate": 5.276381909547739e-06, | |
| "loss": 0.2076, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.9298245614035088, | |
| "grad_norm": 4.7421875, | |
| "learning_rate": 5.1507537688442215e-06, | |
| "loss": 0.2429, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 0.935672514619883, | |
| "grad_norm": 4.375, | |
| "learning_rate": 5.025125628140704e-06, | |
| "loss": 0.1346, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.935672514619883, | |
| "eval_loss": 0.18527910113334656, | |
| "eval_runtime": 129.3266, | |
| "eval_samples_per_second": 3.48, | |
| "eval_steps_per_second": 0.441, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.9415204678362573, | |
| "grad_norm": 4.875, | |
| "learning_rate": 4.899497487437186e-06, | |
| "loss": 0.2457, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 3.013671875, | |
| "learning_rate": 4.773869346733668e-06, | |
| "loss": 0.187, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.9532163742690059, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 4.6482412060301506e-06, | |
| "loss": 0.1546, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 0.9590643274853801, | |
| "grad_norm": 2.5, | |
| "learning_rate": 4.522613065326634e-06, | |
| "loss": 0.1484, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.9590643274853801, | |
| "eval_loss": 0.18491099774837494, | |
| "eval_runtime": 126.5822, | |
| "eval_samples_per_second": 3.555, | |
| "eval_steps_per_second": 0.45, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.9649122807017544, | |
| "grad_norm": 5.03515625, | |
| "learning_rate": 4.396984924623116e-06, | |
| "loss": 0.1739, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 0.9707602339181286, | |
| "grad_norm": 7.1015625, | |
| "learning_rate": 4.271356783919598e-06, | |
| "loss": 0.3268, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.9766081871345029, | |
| "grad_norm": 0.397216796875, | |
| "learning_rate": 4.1457286432160804e-06, | |
| "loss": 0.2178, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "grad_norm": 5.00390625, | |
| "learning_rate": 4.020100502512563e-06, | |
| "loss": 0.1217, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "eval_loss": 0.1838068664073944, | |
| "eval_runtime": 127.461, | |
| "eval_samples_per_second": 3.53, | |
| "eval_steps_per_second": 0.447, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.9883040935672515, | |
| "grad_norm": 0.1112060546875, | |
| "learning_rate": 3.894472361809046e-06, | |
| "loss": 0.0894, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 0.9941520467836257, | |
| "grad_norm": 1.748046875, | |
| "learning_rate": 3.7688442211055276e-06, | |
| "loss": 0.1733, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.53125, | |
| "learning_rate": 3.64321608040201e-06, | |
| "loss": 0.1281, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 1.0058479532163742, | |
| "grad_norm": 0.9111328125, | |
| "learning_rate": 3.5175879396984926e-06, | |
| "loss": 0.0669, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.0058479532163742, | |
| "eval_loss": 0.18437370657920837, | |
| "eval_runtime": 128.6228, | |
| "eval_samples_per_second": 3.499, | |
| "eval_steps_per_second": 0.443, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.0116959064327484, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 3.3919597989949752e-06, | |
| "loss": 0.0748, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 1.0175438596491229, | |
| "grad_norm": 7.703125, | |
| "learning_rate": 3.2663316582914575e-06, | |
| "loss": 0.071, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 1.023391812865497, | |
| "grad_norm": 8.9765625, | |
| "learning_rate": 3.1407035175879398e-06, | |
| "loss": 0.1588, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 1.0292397660818713, | |
| "grad_norm": 0.38427734375, | |
| "learning_rate": 3.015075376884422e-06, | |
| "loss": 0.1292, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.0292397660818713, | |
| "eval_loss": 0.18767422437667847, | |
| "eval_runtime": 126.9252, | |
| "eval_samples_per_second": 3.545, | |
| "eval_steps_per_second": 0.449, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.0350877192982457, | |
| "grad_norm": 6.4375, | |
| "learning_rate": 2.8894472361809047e-06, | |
| "loss": 0.1419, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 1.04093567251462, | |
| "grad_norm": 13.3515625, | |
| "learning_rate": 2.7638190954773874e-06, | |
| "loss": 0.084, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 1.0467836257309941, | |
| "grad_norm": 9.203125, | |
| "learning_rate": 2.6381909547738696e-06, | |
| "loss": 0.072, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 2.439453125, | |
| "learning_rate": 2.512562814070352e-06, | |
| "loss": 0.1106, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "eval_loss": 0.18756501376628876, | |
| "eval_runtime": 128.4204, | |
| "eval_samples_per_second": 3.504, | |
| "eval_steps_per_second": 0.444, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0584795321637426, | |
| "grad_norm": 4.671875, | |
| "learning_rate": 2.386934673366834e-06, | |
| "loss": 0.1226, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 1.064327485380117, | |
| "grad_norm": 2.62890625, | |
| "learning_rate": 2.261306532663317e-06, | |
| "loss": 0.123, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 1.0701754385964912, | |
| "grad_norm": 16.640625, | |
| "learning_rate": 2.135678391959799e-06, | |
| "loss": 0.0593, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 1.0760233918128654, | |
| "grad_norm": 6.8828125, | |
| "learning_rate": 2.0100502512562813e-06, | |
| "loss": 0.0828, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.0760233918128654, | |
| "eval_loss": 0.1875353455543518, | |
| "eval_runtime": 126.5483, | |
| "eval_samples_per_second": 3.556, | |
| "eval_steps_per_second": 0.45, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.0818713450292399, | |
| "grad_norm": 4.55078125, | |
| "learning_rate": 1.8844221105527638e-06, | |
| "loss": 0.1481, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 1.087719298245614, | |
| "grad_norm": 3.6796875, | |
| "learning_rate": 1.7587939698492463e-06, | |
| "loss": 0.1275, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 1.0935672514619883, | |
| "grad_norm": 6.02734375, | |
| "learning_rate": 1.6331658291457288e-06, | |
| "loss": 0.1274, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 1.0994152046783625, | |
| "grad_norm": 1.4111328125, | |
| "learning_rate": 1.507537688442211e-06, | |
| "loss": 0.0485, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.0994152046783625, | |
| "eval_loss": 0.18709704279899597, | |
| "eval_runtime": 129.1721, | |
| "eval_samples_per_second": 3.484, | |
| "eval_steps_per_second": 0.441, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.1052631578947367, | |
| "grad_norm": 0.74853515625, | |
| "learning_rate": 1.3819095477386937e-06, | |
| "loss": 0.0483, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 1.044921875, | |
| "learning_rate": 1.256281407035176e-06, | |
| "loss": 0.0799, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 1.1169590643274854, | |
| "grad_norm": 0.171142578125, | |
| "learning_rate": 1.1306532663316584e-06, | |
| "loss": 0.1273, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 1.1228070175438596, | |
| "grad_norm": 0.346923828125, | |
| "learning_rate": 1.0050251256281407e-06, | |
| "loss": 0.0624, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.1228070175438596, | |
| "eval_loss": 0.1874168962240219, | |
| "eval_runtime": 128.2151, | |
| "eval_samples_per_second": 3.51, | |
| "eval_steps_per_second": 0.445, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.128654970760234, | |
| "grad_norm": 0.1552734375, | |
| "learning_rate": 8.793969849246231e-07, | |
| "loss": 0.1017, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 1.1345029239766082, | |
| "grad_norm": 1.619140625, | |
| "learning_rate": 7.537688442211055e-07, | |
| "loss": 0.1556, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 1.1403508771929824, | |
| "grad_norm": 1.958984375, | |
| "learning_rate": 6.28140703517588e-07, | |
| "loss": 0.1113, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 1.1461988304093567, | |
| "grad_norm": 5.375, | |
| "learning_rate": 5.025125628140703e-07, | |
| "loss": 0.0895, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.1461988304093567, | |
| "eval_loss": 0.1870710253715515, | |
| "eval_runtime": 126.862, | |
| "eval_samples_per_second": 3.547, | |
| "eval_steps_per_second": 0.449, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.1520467836257309, | |
| "grad_norm": 5.984375, | |
| "learning_rate": 3.7688442211055275e-07, | |
| "loss": 0.1076, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 1.1578947368421053, | |
| "grad_norm": 1.9619140625, | |
| "learning_rate": 2.5125628140703517e-07, | |
| "loss": 0.1024, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 1.1637426900584795, | |
| "grad_norm": 6.17578125, | |
| "learning_rate": 1.2562814070351758e-07, | |
| "loss": 0.0698, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 1.1695906432748537, | |
| "grad_norm": 7.06640625, | |
| "learning_rate": 0.0, | |
| "loss": 0.1, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.1695906432748537, | |
| "eval_loss": 0.1871223896741867, | |
| "eval_runtime": 128.9773, | |
| "eval_samples_per_second": 3.489, | |
| "eval_steps_per_second": 0.442, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.1695906432748537, | |
| "step": 5000, | |
| "total_flos": 2.536648947447456e+17, | |
| "train_loss": 0.27304353165626527, | |
| "train_runtime": 21409.3382, | |
| "train_samples_per_second": 0.934, | |
| "train_steps_per_second": 0.234 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.536648947447456e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |