| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 6.887116432189941, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 1.6635, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.9456449747085571, | |
| "eval_runtime": 5.3786, | |
| "eval_samples_per_second": 29.004, | |
| "eval_steps_per_second": 1.487, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 8.283559799194336, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.8201, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.7792918682098389, | |
| "eval_runtime": 5.331, | |
| "eval_samples_per_second": 29.263, | |
| "eval_steps_per_second": 1.501, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 2.8983874320983887, | |
| "learning_rate": 4.2857142857142855e-06, | |
| "loss": 0.7613, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.7527902722358704, | |
| "eval_runtime": 5.3406, | |
| "eval_samples_per_second": 29.21, | |
| "eval_steps_per_second": 1.498, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 3.2681150436401367, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.7325, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.7382247447967529, | |
| "eval_runtime": 5.3526, | |
| "eval_samples_per_second": 29.145, | |
| "eval_steps_per_second": 1.495, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.064232587814331, | |
| "learning_rate": 7.1428571428571436e-06, | |
| "loss": 0.7304, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.7373641729354858, | |
| "eval_runtime": 5.3943, | |
| "eval_samples_per_second": 28.92, | |
| "eval_steps_per_second": 1.483, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 3.030766248703003, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.732, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.7380843758583069, | |
| "eval_runtime": 5.3592, | |
| "eval_samples_per_second": 29.109, | |
| "eval_steps_per_second": 1.493, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 2.9495913982391357, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7425, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.7450628280639648, | |
| "eval_runtime": 5.3499, | |
| "eval_samples_per_second": 29.159, | |
| "eval_steps_per_second": 1.495, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.047274351119995, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 0.727, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.7441740036010742, | |
| "eval_runtime": 5.3431, | |
| "eval_samples_per_second": 29.197, | |
| "eval_steps_per_second": 1.497, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 2.082712411880493, | |
| "learning_rate": 1.2857142857142859e-05, | |
| "loss": 0.7059, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.7533753514289856, | |
| "eval_runtime": 5.3551, | |
| "eval_samples_per_second": 29.131, | |
| "eval_steps_per_second": 1.494, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.9532545804977417, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 0.7125, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.7590879797935486, | |
| "eval_runtime": 5.3446, | |
| "eval_samples_per_second": 29.188, | |
| "eval_steps_per_second": 1.497, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 2.6566646099090576, | |
| "learning_rate": 1.5714285714285715e-05, | |
| "loss": 0.751, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.7629729509353638, | |
| "eval_runtime": 5.3745, | |
| "eval_samples_per_second": 29.026, | |
| "eval_steps_per_second": 1.489, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.752972960472107, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 0.7929, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.7665635943412781, | |
| "eval_runtime": 5.3696, | |
| "eval_samples_per_second": 29.052, | |
| "eval_steps_per_second": 1.49, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.8425724506378174, | |
| "learning_rate": 1.8571428571428575e-05, | |
| "loss": 0.7803, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.7683877944946289, | |
| "eval_runtime": 5.3553, | |
| "eval_samples_per_second": 29.13, | |
| "eval_steps_per_second": 1.494, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.2257883548736572, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7182, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.7802882790565491, | |
| "eval_runtime": 5.3398, | |
| "eval_samples_per_second": 29.214, | |
| "eval_steps_per_second": 1.498, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.124093532562256, | |
| "learning_rate": 1.9996891820008165e-05, | |
| "loss": 0.7318, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.7831437587738037, | |
| "eval_runtime": 5.3755, | |
| "eval_samples_per_second": 29.02, | |
| "eval_steps_per_second": 1.488, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.1694772243499756, | |
| "learning_rate": 1.9987569212189224e-05, | |
| "loss": 0.7509, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.7850997447967529, | |
| "eval_runtime": 5.3457, | |
| "eval_samples_per_second": 29.183, | |
| "eval_steps_per_second": 1.497, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.7642978429794312, | |
| "learning_rate": 1.9972037971811802e-05, | |
| "loss": 0.7617, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.78228759765625, | |
| "eval_runtime": 5.3474, | |
| "eval_samples_per_second": 29.173, | |
| "eval_steps_per_second": 1.496, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.7806785106658936, | |
| "learning_rate": 1.9950307753654016e-05, | |
| "loss": 0.7865, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.784203827381134, | |
| "eval_runtime": 5.3402, | |
| "eval_samples_per_second": 29.212, | |
| "eval_steps_per_second": 1.498, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 2.1907761096954346, | |
| "learning_rate": 1.9922392066001724e-05, | |
| "loss": 0.7619, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.7823160886764526, | |
| "eval_runtime": 5.3429, | |
| "eval_samples_per_second": 29.197, | |
| "eval_steps_per_second": 1.497, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.591108798980713, | |
| "learning_rate": 1.9888308262251286e-05, | |
| "loss": 0.7799, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7811964154243469, | |
| "eval_runtime": 5.3406, | |
| "eval_samples_per_second": 29.21, | |
| "eval_steps_per_second": 1.498, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.7470197677612305, | |
| "learning_rate": 1.9848077530122083e-05, | |
| "loss": 0.4817, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 0.8039106726646423, | |
| "eval_runtime": 5.3508, | |
| "eval_samples_per_second": 29.154, | |
| "eval_steps_per_second": 1.495, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.3562027215957642, | |
| "learning_rate": 1.9801724878485438e-05, | |
| "loss": 0.4391, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.8185109496116638, | |
| "eval_runtime": 5.3553, | |
| "eval_samples_per_second": 29.13, | |
| "eval_steps_per_second": 1.494, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 2.015491247177124, | |
| "learning_rate": 1.9749279121818235e-05, | |
| "loss": 0.4847, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 0.8041576743125916, | |
| "eval_runtime": 5.3543, | |
| "eval_samples_per_second": 29.135, | |
| "eval_steps_per_second": 1.494, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.7701632976531982, | |
| "learning_rate": 1.969077286229078e-05, | |
| "loss": 0.4489, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.8140569925308228, | |
| "eval_runtime": 5.3403, | |
| "eval_samples_per_second": 29.212, | |
| "eval_steps_per_second": 1.498, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.6276088953018188, | |
| "learning_rate": 1.962624246950012e-05, | |
| "loss": 0.4593, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 0.8152586221694946, | |
| "eval_runtime": 5.3674, | |
| "eval_samples_per_second": 29.064, | |
| "eval_steps_per_second": 1.49, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 1.7220439910888672, | |
| "learning_rate": 1.955572805786141e-05, | |
| "loss": 0.4547, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.8013993501663208, | |
| "eval_runtime": 5.3412, | |
| "eval_samples_per_second": 29.207, | |
| "eval_steps_per_second": 1.498, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 1.9381102323532104, | |
| "learning_rate": 1.947927346167132e-05, | |
| "loss": 0.4732, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_loss": 0.8070553541183472, | |
| "eval_runtime": 5.3586, | |
| "eval_samples_per_second": 29.112, | |
| "eval_steps_per_second": 1.493, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.3193718194961548, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.4751, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 0.8212087750434875, | |
| "eval_runtime": 5.3584, | |
| "eval_samples_per_second": 29.113, | |
| "eval_steps_per_second": 1.493, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 1.4072144031524658, | |
| "learning_rate": 1.9308737486442045e-05, | |
| "loss": 0.4711, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.8135795593261719, | |
| "eval_runtime": 5.3462, | |
| "eval_samples_per_second": 29.18, | |
| "eval_steps_per_second": 1.496, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.503187656402588, | |
| "learning_rate": 1.921476211870408e-05, | |
| "loss": 0.4486, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 0.8082787990570068, | |
| "eval_runtime": 5.356, | |
| "eval_samples_per_second": 29.126, | |
| "eval_steps_per_second": 1.494, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.4871103763580322, | |
| "learning_rate": 1.9115058523116734e-05, | |
| "loss": 0.458, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 0.8071246147155762, | |
| "eval_runtime": 5.5657, | |
| "eval_samples_per_second": 28.029, | |
| "eval_steps_per_second": 1.437, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.1439411640167236, | |
| "learning_rate": 1.900968867902419e-05, | |
| "loss": 0.4716, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.8042948246002197, | |
| "eval_runtime": 5.3482, | |
| "eval_samples_per_second": 29.169, | |
| "eval_steps_per_second": 1.496, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 1.5177081823349, | |
| "learning_rate": 1.8898718088114688e-05, | |
| "loss": 0.478, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_loss": 0.8044167757034302, | |
| "eval_runtime": 5.3573, | |
| "eval_samples_per_second": 29.119, | |
| "eval_steps_per_second": 1.493, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 1.6844738721847534, | |
| "learning_rate": 1.8782215733702286e-05, | |
| "loss": 0.5012, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 0.7973454594612122, | |
| "eval_runtime": 5.3529, | |
| "eval_samples_per_second": 29.143, | |
| "eval_steps_per_second": 1.495, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 2.1086294651031494, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 0.4798, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 0.796762228012085, | |
| "eval_runtime": 5.3528, | |
| "eval_samples_per_second": 29.144, | |
| "eval_steps_per_second": 1.495, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.8831372261047363, | |
| "learning_rate": 1.8532908816321557e-05, | |
| "loss": 0.4658, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 0.8035383224487305, | |
| "eval_runtime": 5.3427, | |
| "eval_samples_per_second": 29.199, | |
| "eval_steps_per_second": 1.497, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 1.7962303161621094, | |
| "learning_rate": 1.8400259231507716e-05, | |
| "loss": 0.4761, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 0.8024702072143555, | |
| "eval_runtime": 5.3585, | |
| "eval_samples_per_second": 29.113, | |
| "eval_steps_per_second": 1.493, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.5731513500213623, | |
| "learning_rate": 1.826238774315995e-05, | |
| "loss": 0.5029, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 0.8045446276664734, | |
| "eval_runtime": 5.3667, | |
| "eval_samples_per_second": 29.068, | |
| "eval_steps_per_second": 1.491, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.6794695854187012, | |
| "learning_rate": 1.811938005715857e-05, | |
| "loss": 0.4587, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_loss": 0.8037970066070557, | |
| "eval_runtime": 5.3483, | |
| "eval_samples_per_second": 29.168, | |
| "eval_steps_per_second": 1.496, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.8006577491760254, | |
| "learning_rate": 1.7971325072229227e-05, | |
| "loss": 0.487, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.806293785572052, | |
| "eval_runtime": 5.3609, | |
| "eval_samples_per_second": 29.1, | |
| "eval_steps_per_second": 1.492, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 1.353041648864746, | |
| "learning_rate": 1.78183148246803e-05, | |
| "loss": 0.228, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.8969957232475281, | |
| "eval_runtime": 5.3463, | |
| "eval_samples_per_second": 29.179, | |
| "eval_steps_per_second": 1.496, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 1.695069432258606, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.238, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_loss": 0.8847584128379822, | |
| "eval_runtime": 5.3661, | |
| "eval_samples_per_second": 29.072, | |
| "eval_steps_per_second": 1.491, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 1.4685120582580566, | |
| "learning_rate": 1.7497812029677344e-05, | |
| "loss": 0.2289, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "eval_loss": 0.9000833034515381, | |
| "eval_runtime": 5.3374, | |
| "eval_samples_per_second": 29.228, | |
| "eval_steps_per_second": 1.499, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 1.1362169981002808, | |
| "learning_rate": 1.7330518718298263e-05, | |
| "loss": 0.228, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_loss": 0.9024194478988647, | |
| "eval_runtime": 5.3582, | |
| "eval_samples_per_second": 29.114, | |
| "eval_steps_per_second": 1.493, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 1.252875804901123, | |
| "learning_rate": 1.7158668492597186e-05, | |
| "loss": 0.2262, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 0.9039773941040039, | |
| "eval_runtime": 5.3439, | |
| "eval_samples_per_second": 29.192, | |
| "eval_steps_per_second": 1.497, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 1.2293896675109863, | |
| "learning_rate": 1.698236818086073e-05, | |
| "loss": 0.2401, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_loss": 0.9025784134864807, | |
| "eval_runtime": 5.3641, | |
| "eval_samples_per_second": 29.082, | |
| "eval_steps_per_second": 1.491, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 1.5813263654708862, | |
| "learning_rate": 1.6801727377709195e-05, | |
| "loss": 0.2397, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 0.9166079759597778, | |
| "eval_runtime": 5.3804, | |
| "eval_samples_per_second": 28.994, | |
| "eval_steps_per_second": 1.487, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.4593902826309204, | |
| "learning_rate": 1.6616858375968596e-05, | |
| "loss": 0.2341, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.9137348532676697, | |
| "eval_runtime": 5.3708, | |
| "eval_samples_per_second": 29.046, | |
| "eval_steps_per_second": 1.49, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 1.8817037343978882, | |
| "learning_rate": 1.6427876096865394e-05, | |
| "loss": 0.2434, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_loss": 0.9133093357086182, | |
| "eval_runtime": 5.353, | |
| "eval_samples_per_second": 29.143, | |
| "eval_steps_per_second": 1.494, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.1660922765731812, | |
| "learning_rate": 1.6234898018587336e-05, | |
| "loss": 0.238, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.9203407168388367, | |
| "eval_runtime": 5.3478, | |
| "eval_samples_per_second": 29.171, | |
| "eval_steps_per_second": 1.496, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 1.1189944744110107, | |
| "learning_rate": 1.6038044103254775e-05, | |
| "loss": 0.2555, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.9188438057899475, | |
| "eval_runtime": 5.3606, | |
| "eval_samples_per_second": 29.101, | |
| "eval_steps_per_second": 1.492, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.388344407081604, | |
| "learning_rate": 1.5837436722347902e-05, | |
| "loss": 0.245, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_loss": 0.9264757037162781, | |
| "eval_runtime": 5.3585, | |
| "eval_samples_per_second": 29.112, | |
| "eval_steps_per_second": 1.493, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 1.4485737085342407, | |
| "learning_rate": 1.563320058063622e-05, | |
| "loss": 0.2443, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_loss": 0.9161367416381836, | |
| "eval_runtime": 5.3818, | |
| "eval_samples_per_second": 28.986, | |
| "eval_steps_per_second": 1.486, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 1.536858081817627, | |
| "learning_rate": 1.5425462638657597e-05, | |
| "loss": 0.2377, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 0.919867992401123, | |
| "eval_runtime": 5.3578, | |
| "eval_samples_per_second": 29.116, | |
| "eval_steps_per_second": 1.493, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 1.163506031036377, | |
| "learning_rate": 1.5214352033794981e-05, | |
| "loss": 0.2429, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.9072566628456116, | |
| "eval_runtime": 5.362, | |
| "eval_samples_per_second": 29.094, | |
| "eval_steps_per_second": 1.492, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.5095975399017334, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.2437, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 0.9278251528739929, | |
| "eval_runtime": 5.3835, | |
| "eval_samples_per_second": 28.977, | |
| "eval_steps_per_second": 1.486, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 1.2110488414764404, | |
| "learning_rate": 1.4782539786213184e-05, | |
| "loss": 0.2535, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_loss": 0.9261165261268616, | |
| "eval_runtime": 5.3818, | |
| "eval_samples_per_second": 28.987, | |
| "eval_steps_per_second": 1.486, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 1.4795968532562256, | |
| "learning_rate": 1.4562106573531632e-05, | |
| "loss": 0.2416, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.9134149551391602, | |
| "eval_runtime": 5.3707, | |
| "eval_samples_per_second": 29.046, | |
| "eval_steps_per_second": 1.49, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 1.4406273365020752, | |
| "learning_rate": 1.4338837391175582e-05, | |
| "loss": 0.2455, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_loss": 0.911788821220398, | |
| "eval_runtime": 5.3593, | |
| "eval_samples_per_second": 29.108, | |
| "eval_steps_per_second": 1.493, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.336966633796692, | |
| "learning_rate": 1.4112871031306118e-05, | |
| "loss": 0.2367, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.9113795161247253, | |
| "eval_runtime": 5.3608, | |
| "eval_samples_per_second": 29.1, | |
| "eval_steps_per_second": 1.492, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 7000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 200, | |
| "total_flos": 2.7565823700343194e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |