| { | |
| "best_metric": 1.019782543182373, | |
| "best_model_checkpoint": "/scratch/czm5kz/llama2-13b_32_1_0.0003_sequential_RANDOM_25pct/checkpoint-1220", | |
| "epoch": 0.9975062344139651, | |
| "eval_steps": 20, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.6291937232017517, | |
| "learning_rate": 0.0002989308624376336, | |
| "loss": 2.4791, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.013960838317871, | |
| "learning_rate": 0.00029786172487526725, | |
| "loss": 2.2353, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 5.220999240875244, | |
| "learning_rate": 0.0002967925873129009, | |
| "loss": 2.1904, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8855465650558472, | |
| "learning_rate": 0.00029572344975053457, | |
| "loss": 1.9672, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 1.7276546955108643, | |
| "eval_runtime": 428.596, | |
| "eval_samples_per_second": 26.199, | |
| "eval_steps_per_second": 3.276, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.5181469917297363, | |
| "learning_rate": 0.00029465431218816815, | |
| "loss": 1.8267, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.45906826853752136, | |
| "learning_rate": 0.00029358517462580184, | |
| "loss": 1.7535, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.34283947944641113, | |
| "learning_rate": 0.0002925160370634355, | |
| "loss": 1.6963, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.4426736533641815, | |
| "learning_rate": 0.0002914468995010691, | |
| "loss": 1.6877, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 1.5335692167282104, | |
| "eval_runtime": 429.6602, | |
| "eval_samples_per_second": 26.135, | |
| "eval_steps_per_second": 3.268, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.35644304752349854, | |
| "learning_rate": 0.00029037776193870275, | |
| "loss": 1.6777, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.44168591499328613, | |
| "learning_rate": 0.0002893086243763364, | |
| "loss": 1.6301, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3644905388355255, | |
| "learning_rate": 0.00028823948681397, | |
| "loss": 1.7032, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.2797521650791168, | |
| "learning_rate": 0.0002871703492516037, | |
| "loss": 1.6513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 1.5024319887161255, | |
| "eval_runtime": 429.903, | |
| "eval_samples_per_second": 26.12, | |
| "eval_steps_per_second": 3.266, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4262318015098572, | |
| "learning_rate": 0.0002861012116892373, | |
| "loss": 1.6956, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4277687966823578, | |
| "learning_rate": 0.000285032074126871, | |
| "loss": 1.6304, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.34024539589881897, | |
| "learning_rate": 0.0002839629365645046, | |
| "loss": 1.6168, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.3809501826763153, | |
| "learning_rate": 0.00028289379900213826, | |
| "loss": 1.6026, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.4783284664154053, | |
| "eval_runtime": 429.7998, | |
| "eval_samples_per_second": 26.126, | |
| "eval_steps_per_second": 3.267, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.4502013325691223, | |
| "learning_rate": 0.0002818246614397719, | |
| "loss": 1.6441, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5180115103721619, | |
| "learning_rate": 0.00028075552387740553, | |
| "loss": 1.6145, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.2917430102825165, | |
| "learning_rate": 0.00027968638631503917, | |
| "loss": 1.6093, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.3517865538597107, | |
| "learning_rate": 0.0002786172487526728, | |
| "loss": 1.623, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 1.477242112159729, | |
| "eval_runtime": 430.0134, | |
| "eval_samples_per_second": 26.113, | |
| "eval_steps_per_second": 3.265, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.4099072515964508, | |
| "learning_rate": 0.00027754811119030644, | |
| "loss": 1.5882, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.45024898648262024, | |
| "learning_rate": 0.0002764789736279401, | |
| "loss": 1.5881, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.31718018651008606, | |
| "learning_rate": 0.00027540983606557377, | |
| "loss": 1.6424, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.41389569640159607, | |
| "learning_rate": 0.0002743406985032074, | |
| "loss": 1.5927, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 1.4581289291381836, | |
| "eval_runtime": 428.9623, | |
| "eval_samples_per_second": 26.177, | |
| "eval_steps_per_second": 3.273, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.43200165033340454, | |
| "learning_rate": 0.00027327156094084104, | |
| "loss": 1.6016, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.4232298731803894, | |
| "learning_rate": 0.0002722024233784747, | |
| "loss": 1.5898, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.36732131242752075, | |
| "learning_rate": 0.0002711332858161083, | |
| "loss": 1.5221, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.6660999059677124, | |
| "learning_rate": 0.00027006414825374195, | |
| "loss": 1.6149, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.4388762712478638, | |
| "eval_runtime": 430.0242, | |
| "eval_samples_per_second": 26.112, | |
| "eval_steps_per_second": 3.265, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.7961187958717346, | |
| "learning_rate": 0.00026899501069137564, | |
| "loss": 1.576, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.4951170086860657, | |
| "learning_rate": 0.0002679258731290092, | |
| "loss": 1.5753, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.46930018067359924, | |
| "learning_rate": 0.0002668567355666429, | |
| "loss": 1.5677, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.33915072679519653, | |
| "learning_rate": 0.00026578759800427654, | |
| "loss": 1.5954, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 1.426082968711853, | |
| "eval_runtime": 430.4131, | |
| "eval_samples_per_second": 26.089, | |
| "eval_steps_per_second": 3.262, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.37250855565071106, | |
| "learning_rate": 0.0002647184604419102, | |
| "loss": 1.5982, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.4476079046726227, | |
| "learning_rate": 0.0002636493228795438, | |
| "loss": 1.5739, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.6137750744819641, | |
| "learning_rate": 0.00026258018531717745, | |
| "loss": 1.5652, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.49674519896507263, | |
| "learning_rate": 0.0002615110477548111, | |
| "loss": 1.5676, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.407203197479248, | |
| "eval_runtime": 429.3415, | |
| "eval_samples_per_second": 26.154, | |
| "eval_steps_per_second": 3.27, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.3989889621734619, | |
| "learning_rate": 0.0002604419101924447, | |
| "loss": 1.5738, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.5406049489974976, | |
| "learning_rate": 0.00025937277263007836, | |
| "loss": 1.5633, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.605331301689148, | |
| "learning_rate": 0.000258303635067712, | |
| "loss": 1.6135, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.4816053509712219, | |
| "learning_rate": 0.0002572344975053457, | |
| "loss": 1.557, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 1.38784921169281, | |
| "eval_runtime": 429.6602, | |
| "eval_samples_per_second": 26.135, | |
| "eval_steps_per_second": 3.268, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.4018230140209198, | |
| "learning_rate": 0.0002561653599429793, | |
| "loss": 1.516, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.5622389912605286, | |
| "learning_rate": 0.00025509622238061296, | |
| "loss": 1.543, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.5343061685562134, | |
| "learning_rate": 0.0002540270848182466, | |
| "loss": 1.4964, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.6404452919960022, | |
| "learning_rate": 0.00025295794725588023, | |
| "loss": 1.5498, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 1.357541799545288, | |
| "eval_runtime": 431.0081, | |
| "eval_samples_per_second": 26.053, | |
| "eval_steps_per_second": 3.257, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.5417826771736145, | |
| "learning_rate": 0.00025188880969351387, | |
| "loss": 1.5356, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.5132070183753967, | |
| "learning_rate": 0.00025081967213114756, | |
| "loss": 1.4853, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.5137109160423279, | |
| "learning_rate": 0.00024975053456878114, | |
| "loss": 1.4966, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.538129985332489, | |
| "learning_rate": 0.00024868139700641483, | |
| "loss": 1.5015, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 1.30915105342865, | |
| "eval_runtime": 429.5601, | |
| "eval_samples_per_second": 26.141, | |
| "eval_steps_per_second": 3.268, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.7500670552253723, | |
| "learning_rate": 0.00024761225944404847, | |
| "loss": 1.5566, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.8524929285049438, | |
| "learning_rate": 0.0002465431218816821, | |
| "loss": 1.545, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.7134194374084473, | |
| "learning_rate": 0.00024547398431931574, | |
| "loss": 1.4856, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.6981797814369202, | |
| "learning_rate": 0.0002444048467569494, | |
| "loss": 1.5214, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.2537784576416016, | |
| "eval_runtime": 429.6683, | |
| "eval_samples_per_second": 26.134, | |
| "eval_steps_per_second": 3.268, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.6537020802497864, | |
| "learning_rate": 0.000243335709194583, | |
| "loss": 1.4988, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.7282711863517761, | |
| "learning_rate": 0.00024226657163221665, | |
| "loss": 1.4872, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.990155816078186, | |
| "learning_rate": 0.0002411974340698503, | |
| "loss": 1.4584, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.578199028968811, | |
| "learning_rate": 0.00024012829650748392, | |
| "loss": 1.4595, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.2257404327392578, | |
| "eval_runtime": 430.6646, | |
| "eval_samples_per_second": 26.074, | |
| "eval_steps_per_second": 3.26, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.7419201731681824, | |
| "learning_rate": 0.00023905915894511758, | |
| "loss": 1.4881, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.6976954936981201, | |
| "learning_rate": 0.00023799002138275122, | |
| "loss": 1.4743, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.6367508769035339, | |
| "learning_rate": 0.00023692088382038488, | |
| "loss": 1.4695, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.7029184699058533, | |
| "learning_rate": 0.0002358517462580185, | |
| "loss": 1.4245, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 1.1911545991897583, | |
| "eval_runtime": 430.7506, | |
| "eval_samples_per_second": 26.068, | |
| "eval_steps_per_second": 3.259, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.5411133170127869, | |
| "learning_rate": 0.00023478260869565215, | |
| "loss": 1.4176, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.8051818609237671, | |
| "learning_rate": 0.0002337134711332858, | |
| "loss": 1.4264, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.6067179441452026, | |
| "learning_rate": 0.00023264433357091945, | |
| "loss": 1.4669, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.7595986127853394, | |
| "learning_rate": 0.0002315751960085531, | |
| "loss": 1.4295, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 1.1661920547485352, | |
| "eval_runtime": 430.0879, | |
| "eval_samples_per_second": 26.109, | |
| "eval_steps_per_second": 3.264, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.6554045081138611, | |
| "learning_rate": 0.00023050605844618672, | |
| "loss": 1.4938, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.3542299270629883, | |
| "learning_rate": 0.00022943692088382036, | |
| "loss": 1.3983, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.7835495471954346, | |
| "learning_rate": 0.00022836778332145402, | |
| "loss": 1.3958, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.695136547088623, | |
| "learning_rate": 0.00022729864575908766, | |
| "loss": 1.45, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.1369729042053223, | |
| "eval_runtime": 429.726, | |
| "eval_samples_per_second": 26.131, | |
| "eval_steps_per_second": 3.267, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.8016501069068909, | |
| "learning_rate": 0.00022622950819672127, | |
| "loss": 1.3158, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.617300271987915, | |
| "learning_rate": 0.00022516037063435493, | |
| "loss": 1.4326, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.0931488275527954, | |
| "learning_rate": 0.00022409123307198857, | |
| "loss": 1.4018, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.6332559585571289, | |
| "learning_rate": 0.00022302209550962223, | |
| "loss": 1.3883, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 1.1649107933044434, | |
| "eval_runtime": 429.708, | |
| "eval_samples_per_second": 26.132, | |
| "eval_steps_per_second": 3.267, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.4458891153335571, | |
| "learning_rate": 0.00022195295794725584, | |
| "loss": 1.4136, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.1108673810958862, | |
| "learning_rate": 0.0002208838203848895, | |
| "loss": 1.4037, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.50286465883255, | |
| "learning_rate": 0.00021981468282252314, | |
| "loss": 1.4646, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.6288882493972778, | |
| "learning_rate": 0.0002187455452601568, | |
| "loss": 1.377, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.1202141046524048, | |
| "eval_runtime": 430.2691, | |
| "eval_samples_per_second": 26.098, | |
| "eval_steps_per_second": 3.263, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.6638579368591309, | |
| "learning_rate": 0.0002176764076977904, | |
| "loss": 1.3928, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.6448540091514587, | |
| "learning_rate": 0.00021660727013542407, | |
| "loss": 1.3941, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.5767375826835632, | |
| "learning_rate": 0.0002155381325730577, | |
| "loss": 1.3517, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.5582290887832642, | |
| "learning_rate": 0.00021446899501069137, | |
| "loss": 1.4041, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.0965207815170288, | |
| "eval_runtime": 429.9294, | |
| "eval_samples_per_second": 26.118, | |
| "eval_steps_per_second": 3.266, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.8463812470436096, | |
| "learning_rate": 0.00021339985744832498, | |
| "loss": 1.3467, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.5267343521118164, | |
| "learning_rate": 0.00021233071988595865, | |
| "loss": 1.3615, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.7553706169128418, | |
| "learning_rate": 0.00021126158232359228, | |
| "loss": 1.3438, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.7279307842254639, | |
| "learning_rate": 0.00021019244476122595, | |
| "loss": 1.4185, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 1.1018279790878296, | |
| "eval_runtime": 429.0802, | |
| "eval_samples_per_second": 26.17, | |
| "eval_steps_per_second": 3.272, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.7685341835021973, | |
| "learning_rate": 0.00020912330719885958, | |
| "loss": 1.4186, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.7275740504264832, | |
| "learning_rate": 0.0002080541696364932, | |
| "loss": 1.3576, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.7372286319732666, | |
| "learning_rate": 0.00020698503207412685, | |
| "loss": 1.3902, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.8976317644119263, | |
| "learning_rate": 0.0002059158945117605, | |
| "loss": 1.404, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 1.09088933467865, | |
| "eval_runtime": 429.9108, | |
| "eval_samples_per_second": 26.119, | |
| "eval_steps_per_second": 3.266, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.7758693695068359, | |
| "learning_rate": 0.00020484675694939415, | |
| "loss": 1.3686, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.8729623556137085, | |
| "learning_rate": 0.00020377761938702776, | |
| "loss": 1.3923, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.7426387667655945, | |
| "learning_rate": 0.00020270848182466143, | |
| "loss": 1.3244, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5695792436599731, | |
| "learning_rate": 0.00020163934426229506, | |
| "loss": 1.3677, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 1.0908492803573608, | |
| "eval_runtime": 428.9402, | |
| "eval_samples_per_second": 26.178, | |
| "eval_steps_per_second": 3.273, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5147937536239624, | |
| "learning_rate": 0.00020057020669992872, | |
| "loss": 1.3643, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.5052825212478638, | |
| "learning_rate": 0.00019950106913756233, | |
| "loss": 1.3665, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.4269680976867676, | |
| "learning_rate": 0.000198431931575196, | |
| "loss": 1.3191, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.49817460775375366, | |
| "learning_rate": 0.00019736279401282963, | |
| "loss": 1.3423, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 1.081033706665039, | |
| "eval_runtime": 430.1153, | |
| "eval_samples_per_second": 26.107, | |
| "eval_steps_per_second": 3.264, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.5547080636024475, | |
| "learning_rate": 0.0001962936564504633, | |
| "loss": 1.3562, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.851958692073822, | |
| "learning_rate": 0.0001952245188880969, | |
| "loss": 1.4167, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.8551843166351318, | |
| "learning_rate": 0.00019415538132573057, | |
| "loss": 1.4025, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.6752102971076965, | |
| "learning_rate": 0.0001930862437633642, | |
| "loss": 1.3144, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.0697931051254272, | |
| "eval_runtime": 430.0744, | |
| "eval_samples_per_second": 26.109, | |
| "eval_steps_per_second": 3.265, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.5209426283836365, | |
| "learning_rate": 0.00019201710620099787, | |
| "loss": 1.405, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.5636367201805115, | |
| "learning_rate": 0.00019094796863863148, | |
| "loss": 1.3456, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.5399631857872009, | |
| "learning_rate": 0.0001898788310762651, | |
| "loss": 1.3752, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.6341372728347778, | |
| "learning_rate": 0.00018880969351389878, | |
| "loss": 1.3685, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 1.0655372142791748, | |
| "eval_runtime": 430.0079, | |
| "eval_samples_per_second": 26.113, | |
| "eval_steps_per_second": 3.265, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.018846869468689, | |
| "learning_rate": 0.0001877405559515324, | |
| "loss": 1.3351, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.6910350322723389, | |
| "learning_rate": 0.00018667141838916605, | |
| "loss": 1.3409, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 1.3985555171966553, | |
| "learning_rate": 0.00018560228082679968, | |
| "loss": 1.3792, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.9939374327659607, | |
| "learning_rate": 0.00018453314326443335, | |
| "loss": 1.341, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 1.0706018209457397, | |
| "eval_runtime": 430.0786, | |
| "eval_samples_per_second": 26.109, | |
| "eval_steps_per_second": 3.265, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.8195542693138123, | |
| "learning_rate": 0.00018346400570206698, | |
| "loss": 1.3279, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.6725456118583679, | |
| "learning_rate": 0.00018239486813970065, | |
| "loss": 1.3687, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5679619908332825, | |
| "learning_rate": 0.00018132573057733425, | |
| "loss": 1.3672, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5750882029533386, | |
| "learning_rate": 0.00018025659301496792, | |
| "loss": 1.3077, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.0635442733764648, | |
| "eval_runtime": 430.3677, | |
| "eval_samples_per_second": 26.092, | |
| "eval_steps_per_second": 3.262, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.6369686722755432, | |
| "learning_rate": 0.00017918745545260155, | |
| "loss": 1.3236, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.5866047143936157, | |
| "learning_rate": 0.00017811831789023522, | |
| "loss": 1.3253, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.5465147495269775, | |
| "learning_rate": 0.00017704918032786883, | |
| "loss": 1.3344, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.6831965446472168, | |
| "learning_rate": 0.0001759800427655025, | |
| "loss": 1.3546, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 1.0727200508117676, | |
| "eval_runtime": 430.2107, | |
| "eval_samples_per_second": 26.101, | |
| "eval_steps_per_second": 3.264, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.9631245136260986, | |
| "learning_rate": 0.00017491090520313613, | |
| "loss": 1.3735, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.8638582825660706, | |
| "learning_rate": 0.0001738417676407698, | |
| "loss": 1.3783, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.5413733720779419, | |
| "learning_rate": 0.0001727726300784034, | |
| "loss": 1.3831, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.5948848128318787, | |
| "learning_rate": 0.00017170349251603703, | |
| "loss": 1.3589, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 1.0601730346679688, | |
| "eval_runtime": 429.9557, | |
| "eval_samples_per_second": 26.117, | |
| "eval_steps_per_second": 3.265, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.7511246800422668, | |
| "learning_rate": 0.0001706343549536707, | |
| "loss": 1.3685, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.6298426389694214, | |
| "learning_rate": 0.00016956521739130433, | |
| "loss": 1.309, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.6661334037780762, | |
| "learning_rate": 0.00016849607982893797, | |
| "loss": 1.3337, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.5155322551727295, | |
| "learning_rate": 0.0001674269422665716, | |
| "loss": 1.3756, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.0586997270584106, | |
| "eval_runtime": 430.2762, | |
| "eval_samples_per_second": 26.097, | |
| "eval_steps_per_second": 3.263, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.3205231428146362, | |
| "learning_rate": 0.00016635780470420527, | |
| "loss": 1.3285, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.4283859133720398, | |
| "learning_rate": 0.0001652886671418389, | |
| "loss": 1.3028, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.1601816415786743, | |
| "learning_rate": 0.00016421952957947254, | |
| "loss": 1.3416, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.44518712162971497, | |
| "learning_rate": 0.00016315039201710618, | |
| "loss": 1.3171, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 1.0572543144226074, | |
| "eval_runtime": 429.3198, | |
| "eval_samples_per_second": 26.155, | |
| "eval_steps_per_second": 3.27, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.7820278406143188, | |
| "learning_rate": 0.00016208125445473984, | |
| "loss": 1.3304, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.6720359325408936, | |
| "learning_rate": 0.00016101211689237348, | |
| "loss": 1.3263, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1.0143332481384277, | |
| "learning_rate": 0.00015994297933000714, | |
| "loss": 1.332, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.5054823160171509, | |
| "learning_rate": 0.00015887384176764075, | |
| "loss": 1.2838, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 1.064550518989563, | |
| "eval_runtime": 429.3314, | |
| "eval_samples_per_second": 26.155, | |
| "eval_steps_per_second": 3.27, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1.031082272529602, | |
| "learning_rate": 0.0001578047042052744, | |
| "loss": 1.3094, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.9540282487869263, | |
| "learning_rate": 0.00015673556664290805, | |
| "loss": 1.3874, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.5223524570465088, | |
| "learning_rate": 0.0001556664290805417, | |
| "loss": 1.3256, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.7279605269432068, | |
| "learning_rate": 0.00015459729151817532, | |
| "loss": 1.373, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 1.0548007488250732, | |
| "eval_runtime": 429.2395, | |
| "eval_samples_per_second": 26.16, | |
| "eval_steps_per_second": 3.271, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.5409196019172668, | |
| "learning_rate": 0.00015352815395580896, | |
| "loss": 1.3431, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.5752881765365601, | |
| "learning_rate": 0.00015245901639344262, | |
| "loss": 1.3486, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.8059548139572144, | |
| "learning_rate": 0.00015138987883107623, | |
| "loss": 1.3471, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.509373664855957, | |
| "learning_rate": 0.0001503207412687099, | |
| "loss": 1.3501, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.0632661581039429, | |
| "eval_runtime": 429.2245, | |
| "eval_samples_per_second": 26.161, | |
| "eval_steps_per_second": 3.271, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.6135790944099426, | |
| "learning_rate": 0.00014925160370634355, | |
| "loss": 1.3213, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.6826092600822449, | |
| "learning_rate": 0.0001481824661439772, | |
| "loss": 1.3623, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.6348379850387573, | |
| "learning_rate": 0.00014711332858161083, | |
| "loss": 1.3135, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.7005529999732971, | |
| "learning_rate": 0.00014604419101924446, | |
| "loss": 1.3419, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.0513877868652344, | |
| "eval_runtime": 430.4426, | |
| "eval_samples_per_second": 26.087, | |
| "eval_steps_per_second": 3.262, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.5816593170166016, | |
| "learning_rate": 0.0001449750534568781, | |
| "loss": 1.3094, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.6071094274520874, | |
| "learning_rate": 0.00014390591589451173, | |
| "loss": 1.2911, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.0864328145980835, | |
| "learning_rate": 0.0001428367783321454, | |
| "loss": 1.3191, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.8440472483634949, | |
| "learning_rate": 0.00014176764076977903, | |
| "loss": 1.2931, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.0493059158325195, | |
| "eval_runtime": 430.1744, | |
| "eval_samples_per_second": 26.103, | |
| "eval_steps_per_second": 3.264, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.238255262374878, | |
| "learning_rate": 0.00014069850320741267, | |
| "loss": 1.2751, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.5711309909820557, | |
| "learning_rate": 0.0001396293656450463, | |
| "loss": 1.3271, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.7740243077278137, | |
| "learning_rate": 0.00013856022808267997, | |
| "loss": 1.3454, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.4949895441532135, | |
| "learning_rate": 0.0001374910905203136, | |
| "loss": 1.3521, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 1.0512796640396118, | |
| "eval_runtime": 429.3325, | |
| "eval_samples_per_second": 26.155, | |
| "eval_steps_per_second": 3.27, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.8579160571098328, | |
| "learning_rate": 0.00013642195295794724, | |
| "loss": 1.2972, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.8241709470748901, | |
| "learning_rate": 0.00013535281539558088, | |
| "loss": 1.3556, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.7028602957725525, | |
| "learning_rate": 0.00013428367783321454, | |
| "loss": 1.3457, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5605300068855286, | |
| "learning_rate": 0.00013321454027084818, | |
| "loss": 1.3149, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 1.0340629816055298, | |
| "eval_runtime": 428.4333, | |
| "eval_samples_per_second": 26.209, | |
| "eval_steps_per_second": 3.277, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5241943001747131, | |
| "learning_rate": 0.0001321454027084818, | |
| "loss": 1.3316, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.7008923888206482, | |
| "learning_rate": 0.00013107626514611545, | |
| "loss": 1.3343, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.7116791009902954, | |
| "learning_rate": 0.0001300071275837491, | |
| "loss": 1.3599, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.47471854090690613, | |
| "learning_rate": 0.00012893799002138275, | |
| "loss": 1.3042, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 1.0408192873001099, | |
| "eval_runtime": 430.9133, | |
| "eval_samples_per_second": 26.059, | |
| "eval_steps_per_second": 3.258, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.4862081706523895, | |
| "learning_rate": 0.00012786885245901638, | |
| "loss": 1.3269, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.6292262077331543, | |
| "learning_rate": 0.00012679971489665002, | |
| "loss": 1.3372, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.5607986450195312, | |
| "learning_rate": 0.00012573057733428366, | |
| "loss": 1.3225, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.9970059394836426, | |
| "learning_rate": 0.0001246614397719173, | |
| "loss": 1.3126, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.0392816066741943, | |
| "eval_runtime": 429.9226, | |
| "eval_samples_per_second": 26.119, | |
| "eval_steps_per_second": 3.266, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.5598148703575134, | |
| "learning_rate": 0.00012359230220955095, | |
| "loss": 1.3226, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.7155619859695435, | |
| "learning_rate": 0.0001225231646471846, | |
| "loss": 1.3015, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.5198763012886047, | |
| "learning_rate": 0.00012145402708481824, | |
| "loss": 1.3338, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.6012284755706787, | |
| "learning_rate": 0.00012038488952245188, | |
| "loss": 1.3306, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 1.0436369180679321, | |
| "eval_runtime": 430.0367, | |
| "eval_samples_per_second": 26.112, | |
| "eval_steps_per_second": 3.265, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.7888386249542236, | |
| "learning_rate": 0.00011931575196008553, | |
| "loss": 1.3209, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.598448634147644, | |
| "learning_rate": 0.00011824661439771916, | |
| "loss": 1.351, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.6814375519752502, | |
| "learning_rate": 0.00011717747683535281, | |
| "loss": 1.3323, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.5430750846862793, | |
| "learning_rate": 0.00011610833927298645, | |
| "loss": 1.299, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 1.0388227701187134, | |
| "eval_runtime": 429.5954, | |
| "eval_samples_per_second": 26.139, | |
| "eval_steps_per_second": 3.268, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.0289543867111206, | |
| "learning_rate": 0.0001150392017106201, | |
| "loss": 1.3167, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.8182229399681091, | |
| "learning_rate": 0.00011397006414825373, | |
| "loss": 1.2939, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.5602372288703918, | |
| "learning_rate": 0.00011290092658588738, | |
| "loss": 1.345, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.8015730977058411, | |
| "learning_rate": 0.00011183178902352102, | |
| "loss": 1.3186, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 1.0438414812088013, | |
| "eval_runtime": 430.3302, | |
| "eval_samples_per_second": 26.094, | |
| "eval_steps_per_second": 3.263, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.5967668890953064, | |
| "learning_rate": 0.00011076265146115467, | |
| "loss": 1.3232, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.8282290697097778, | |
| "learning_rate": 0.00010969351389878829, | |
| "loss": 1.2881, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.5095836520195007, | |
| "learning_rate": 0.00010862437633642194, | |
| "loss": 1.3026, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.5845779776573181, | |
| "learning_rate": 0.00010755523877405558, | |
| "loss": 1.385, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 1.0384798049926758, | |
| "eval_runtime": 429.8904, | |
| "eval_samples_per_second": 26.121, | |
| "eval_steps_per_second": 3.266, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.6281318664550781, | |
| "learning_rate": 0.00010648610121168923, | |
| "loss": 1.3167, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.4388660192489624, | |
| "learning_rate": 0.00010541696364932286, | |
| "loss": 1.2848, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.5878056883811951, | |
| "learning_rate": 0.00010434782608695651, | |
| "loss": 1.2798, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.0295441150665283, | |
| "learning_rate": 0.00010327868852459015, | |
| "loss": 1.3578, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.036583662033081, | |
| "eval_runtime": 429.2823, | |
| "eval_samples_per_second": 26.158, | |
| "eval_steps_per_second": 3.271, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.169296383857727, | |
| "learning_rate": 0.0001022095509622238, | |
| "loss": 1.3251, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.8990792632102966, | |
| "learning_rate": 0.00010114041339985743, | |
| "loss": 1.3402, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.5439760088920593, | |
| "learning_rate": 0.00010007127583749108, | |
| "loss": 1.302, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.5038203597068787, | |
| "learning_rate": 9.900213827512472e-05, | |
| "loss": 1.2863, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.0299803018569946, | |
| "eval_runtime": 430.5328, | |
| "eval_samples_per_second": 26.082, | |
| "eval_steps_per_second": 3.261, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.5802446603775024, | |
| "learning_rate": 9.793300071275837e-05, | |
| "loss": 1.2493, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.5156883001327515, | |
| "learning_rate": 9.686386315039202e-05, | |
| "loss": 1.2997, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.7282747030258179, | |
| "learning_rate": 9.579472558802566e-05, | |
| "loss": 1.2893, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.5055095553398132, | |
| "learning_rate": 9.47255880256593e-05, | |
| "loss": 1.2894, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.0320148468017578, | |
| "eval_runtime": 428.6868, | |
| "eval_samples_per_second": 26.194, | |
| "eval_steps_per_second": 3.275, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.7030293345451355, | |
| "learning_rate": 9.365645046329294e-05, | |
| "loss": 1.3313, | |
| "step": 965 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.4963017404079437, | |
| "learning_rate": 9.258731290092659e-05, | |
| "loss": 1.2798, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 1.1872601509094238, | |
| "learning_rate": 9.151817533856021e-05, | |
| "loss": 1.3095, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.01945960521698, | |
| "learning_rate": 9.044903777619385e-05, | |
| "loss": 1.319, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 1.037099838256836, | |
| "eval_runtime": 428.5058, | |
| "eval_samples_per_second": 26.205, | |
| "eval_steps_per_second": 3.277, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.9102843999862671, | |
| "learning_rate": 8.93799002138275e-05, | |
| "loss": 1.2645, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.43511050939559937, | |
| "learning_rate": 8.831076265146115e-05, | |
| "loss": 1.3248, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.866431713104248, | |
| "learning_rate": 8.724162508909478e-05, | |
| "loss": 1.2844, | |
| "step": 995 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.5549300909042358, | |
| "learning_rate": 8.617248752672843e-05, | |
| "loss": 1.2784, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 1.036270260810852, | |
| "eval_runtime": 430.0845, | |
| "eval_samples_per_second": 26.109, | |
| "eval_steps_per_second": 3.264, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.503010094165802, | |
| "learning_rate": 8.510334996436207e-05, | |
| "loss": 1.2949, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.6416985988616943, | |
| "learning_rate": 8.403421240199572e-05, | |
| "loss": 1.3145, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.5773264765739441, | |
| "learning_rate": 8.296507483962936e-05, | |
| "loss": 1.2916, | |
| "step": 1015 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.8444538116455078, | |
| "learning_rate": 8.1895937277263e-05, | |
| "loss": 1.2588, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.02756929397583, | |
| "eval_runtime": 429.6856, | |
| "eval_samples_per_second": 26.133, | |
| "eval_steps_per_second": 3.268, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.6495450735092163, | |
| "learning_rate": 8.082679971489664e-05, | |
| "loss": 1.3093, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.633300244808197, | |
| "learning_rate": 7.975766215253029e-05, | |
| "loss": 1.292, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.5709363222122192, | |
| "learning_rate": 7.868852459016393e-05, | |
| "loss": 1.2976, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.6210110187530518, | |
| "learning_rate": 7.761938702779758e-05, | |
| "loss": 1.3409, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_loss": 1.0304205417633057, | |
| "eval_runtime": 430.053, | |
| "eval_samples_per_second": 26.111, | |
| "eval_steps_per_second": 3.265, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.5457026958465576, | |
| "learning_rate": 7.655024946543121e-05, | |
| "loss": 1.2711, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.5168452858924866, | |
| "learning_rate": 7.548111190306486e-05, | |
| "loss": 1.3493, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.6571021676063538, | |
| "learning_rate": 7.44119743406985e-05, | |
| "loss": 1.3051, | |
| "step": 1055 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.6311302185058594, | |
| "learning_rate": 7.334283677833213e-05, | |
| "loss": 1.2985, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 1.0277342796325684, | |
| "eval_runtime": 429.4921, | |
| "eval_samples_per_second": 26.145, | |
| "eval_steps_per_second": 3.269, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.5146097540855408, | |
| "learning_rate": 7.227369921596578e-05, | |
| "loss": 1.2888, | |
| "step": 1065 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.9732445478439331, | |
| "learning_rate": 7.120456165359942e-05, | |
| "loss": 1.3134, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.126225233078003, | |
| "learning_rate": 7.013542409123307e-05, | |
| "loss": 1.3117, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.6155526041984558, | |
| "learning_rate": 6.90662865288667e-05, | |
| "loss": 1.2754, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_loss": 1.0243170261383057, | |
| "eval_runtime": 429.5325, | |
| "eval_samples_per_second": 26.142, | |
| "eval_steps_per_second": 3.269, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.5756591558456421, | |
| "learning_rate": 6.799714896650034e-05, | |
| "loss": 1.2867, | |
| "step": 1085 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.9171560406684875, | |
| "learning_rate": 6.692801140413399e-05, | |
| "loss": 1.3024, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.0372343063354492, | |
| "learning_rate": 6.585887384176763e-05, | |
| "loss": 1.3004, | |
| "step": 1095 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.5745114684104919, | |
| "learning_rate": 6.478973627940128e-05, | |
| "loss": 1.3079, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 1.0274326801300049, | |
| "eval_runtime": 430.2338, | |
| "eval_samples_per_second": 26.1, | |
| "eval_steps_per_second": 3.263, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.6692591309547424, | |
| "learning_rate": 6.372059871703493e-05, | |
| "loss": 1.3284, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.5350373983383179, | |
| "learning_rate": 6.265146115466856e-05, | |
| "loss": 1.2706, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.5629821419715881, | |
| "learning_rate": 6.158232359230221e-05, | |
| "loss": 1.2898, | |
| "step": 1115 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.6059845685958862, | |
| "learning_rate": 6.051318602993584e-05, | |
| "loss": 1.2614, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.0247998237609863, | |
| "eval_runtime": 430.8367, | |
| "eval_samples_per_second": 26.063, | |
| "eval_steps_per_second": 3.259, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.9147688150405884, | |
| "learning_rate": 5.9444048467569485e-05, | |
| "loss": 1.3293, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.48272547125816345, | |
| "learning_rate": 5.837491090520313e-05, | |
| "loss": 1.3069, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.5300667881965637, | |
| "learning_rate": 5.730577334283677e-05, | |
| "loss": 1.3209, | |
| "step": 1135 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.5542011857032776, | |
| "learning_rate": 5.6236635780470413e-05, | |
| "loss": 1.3116, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 1.027745246887207, | |
| "eval_runtime": 428.9108, | |
| "eval_samples_per_second": 26.18, | |
| "eval_steps_per_second": 3.273, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.6744430661201477, | |
| "learning_rate": 5.5167498218104056e-05, | |
| "loss": 1.2887, | |
| "step": 1145 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.660572350025177, | |
| "learning_rate": 5.40983606557377e-05, | |
| "loss": 1.2296, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.6836673021316528, | |
| "learning_rate": 5.302922309337134e-05, | |
| "loss": 1.2773, | |
| "step": 1155 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.5106320381164551, | |
| "learning_rate": 5.196008553100499e-05, | |
| "loss": 1.3101, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_loss": 1.0228785276412964, | |
| "eval_runtime": 430.0103, | |
| "eval_samples_per_second": 26.113, | |
| "eval_steps_per_second": 3.265, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.5948910117149353, | |
| "learning_rate": 5.089094796863862e-05, | |
| "loss": 1.3074, | |
| "step": 1165 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.6630853414535522, | |
| "learning_rate": 4.9821810406272264e-05, | |
| "loss": 1.3249, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.5019823312759399, | |
| "learning_rate": 4.875267284390591e-05, | |
| "loss": 1.3239, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.524602472782135, | |
| "learning_rate": 4.7683535281539556e-05, | |
| "loss": 1.3213, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.028998613357544, | |
| "eval_runtime": 430.7493, | |
| "eval_samples_per_second": 26.069, | |
| "eval_steps_per_second": 3.259, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.5343472361564636, | |
| "learning_rate": 4.66143977191732e-05, | |
| "loss": 1.3257, | |
| "step": 1185 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.6384676098823547, | |
| "learning_rate": 4.554526015680684e-05, | |
| "loss": 1.3164, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.6373520493507385, | |
| "learning_rate": 4.4476122594440485e-05, | |
| "loss": 1.3331, | |
| "step": 1195 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.9878657460212708, | |
| "learning_rate": 4.340698503207413e-05, | |
| "loss": 1.3003, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 1.0289608240127563, | |
| "eval_runtime": 429.2778, | |
| "eval_samples_per_second": 26.158, | |
| "eval_steps_per_second": 3.271, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.5717479586601257, | |
| "learning_rate": 4.2337847469707764e-05, | |
| "loss": 1.3245, | |
| "step": 1205 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.7836593985557556, | |
| "learning_rate": 4.1268709907341407e-05, | |
| "loss": 1.3023, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.53411465883255, | |
| "learning_rate": 4.019957234497505e-05, | |
| "loss": 1.2955, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.5131638646125793, | |
| "learning_rate": 3.913043478260869e-05, | |
| "loss": 1.251, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.019782543182373, | |
| "eval_runtime": 429.0558, | |
| "eval_samples_per_second": 26.171, | |
| "eval_steps_per_second": 3.272, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.5704027414321899, | |
| "learning_rate": 3.8061297220242335e-05, | |
| "loss": 1.3267, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.44694724678993225, | |
| "learning_rate": 3.699215965787598e-05, | |
| "loss": 1.307, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.5663137435913086, | |
| "learning_rate": 3.592302209550962e-05, | |
| "loss": 1.3299, | |
| "step": 1235 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.7042715549468994, | |
| "learning_rate": 3.485388453314326e-05, | |
| "loss": 1.283, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 1.0289275646209717, | |
| "eval_runtime": 429.5115, | |
| "eval_samples_per_second": 26.144, | |
| "eval_steps_per_second": 3.269, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.6489189267158508, | |
| "learning_rate": 3.3784746970776906e-05, | |
| "loss": 1.3028, | |
| "step": 1245 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.6258445382118225, | |
| "learning_rate": 3.271560940841055e-05, | |
| "loss": 1.2966, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.5730215311050415, | |
| "learning_rate": 3.164647184604419e-05, | |
| "loss": 1.318, | |
| "step": 1255 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.7148025035858154, | |
| "learning_rate": 3.057733428367783e-05, | |
| "loss": 1.263, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 1.0275732278823853, | |
| "eval_runtime": 430.3813, | |
| "eval_samples_per_second": 26.091, | |
| "eval_steps_per_second": 3.262, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.5665017366409302, | |
| "learning_rate": 2.950819672131147e-05, | |
| "loss": 1.2744, | |
| "step": 1265 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.8906866312026978, | |
| "learning_rate": 2.8439059158945114e-05, | |
| "loss": 1.3154, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.7491171956062317, | |
| "learning_rate": 2.736992159657876e-05, | |
| "loss": 1.3158, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.5208207964897156, | |
| "learning_rate": 2.63007840342124e-05, | |
| "loss": 1.2488, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 1.02449369430542, | |
| "eval_runtime": 429.3553, | |
| "eval_samples_per_second": 26.153, | |
| "eval_steps_per_second": 3.27, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.7041420936584473, | |
| "learning_rate": 2.5231646471846042e-05, | |
| "loss": 1.3145, | |
| "step": 1285 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.5100432634353638, | |
| "learning_rate": 2.4162508909479685e-05, | |
| "loss": 1.3209, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.491322785615921, | |
| "learning_rate": 2.3093371347113328e-05, | |
| "loss": 1.2835, | |
| "step": 1295 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.4475841224193573, | |
| "learning_rate": 2.2024233784746968e-05, | |
| "loss": 1.3328, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.0273866653442383, | |
| "eval_runtime": 429.937, | |
| "eval_samples_per_second": 26.118, | |
| "eval_steps_per_second": 3.266, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.4693332612514496, | |
| "learning_rate": 2.095509622238061e-05, | |
| "loss": 1.284, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.5190310478210449, | |
| "learning_rate": 1.9885958660014253e-05, | |
| "loss": 1.2987, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.5245566368103027, | |
| "learning_rate": 1.8816821097647896e-05, | |
| "loss": 1.3144, | |
| "step": 1315 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.6682144403457642, | |
| "learning_rate": 1.774768353528154e-05, | |
| "loss": 1.331, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 1.0257986783981323, | |
| "eval_runtime": 429.8676, | |
| "eval_samples_per_second": 26.122, | |
| "eval_steps_per_second": 3.266, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.8946670293807983, | |
| "learning_rate": 1.6678545972915182e-05, | |
| "loss": 1.307, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.4281274378299713, | |
| "learning_rate": 1.560940841054882e-05, | |
| "loss": 1.316, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.6014120578765869, | |
| "learning_rate": 1.4540270848182466e-05, | |
| "loss": 1.3436, | |
| "step": 1335 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.48019564151763916, | |
| "learning_rate": 1.3471133285816107e-05, | |
| "loss": 1.247, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.026702642440796, | |
| "eval_runtime": 429.7224, | |
| "eval_samples_per_second": 26.131, | |
| "eval_steps_per_second": 3.267, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.6991472244262695, | |
| "learning_rate": 1.240199572344975e-05, | |
| "loss": 1.3156, | |
| "step": 1345 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5677585601806641, | |
| "learning_rate": 1.1332858161083391e-05, | |
| "loss": 1.3139, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.4951375722885132, | |
| "learning_rate": 1.0263720598717034e-05, | |
| "loss": 1.2654, | |
| "step": 1355 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.4676544666290283, | |
| "learning_rate": 9.194583036350677e-06, | |
| "loss": 1.2781, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 1.0250627994537354, | |
| "eval_runtime": 428.9007, | |
| "eval_samples_per_second": 26.181, | |
| "eval_steps_per_second": 3.273, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.5621703267097473, | |
| "learning_rate": 8.12544547398432e-06, | |
| "loss": 1.3214, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.5663436055183411, | |
| "learning_rate": 7.0563079116179615e-06, | |
| "loss": 1.3685, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.6462903022766113, | |
| "learning_rate": 5.9871703492516035e-06, | |
| "loss": 1.3161, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.5171219706535339, | |
| "learning_rate": 4.9180327868852455e-06, | |
| "loss": 1.2707, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 1.0249730348587036, | |
| "eval_runtime": 429.5565, | |
| "eval_samples_per_second": 26.141, | |
| "eval_steps_per_second": 3.268, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.5951040387153625, | |
| "learning_rate": 3.848895224518888e-06, | |
| "loss": 1.3382, | |
| "step": 1385 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.43818196654319763, | |
| "learning_rate": 2.7797576621525303e-06, | |
| "loss": 1.2831, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.6639280319213867, | |
| "learning_rate": 1.7106200997861725e-06, | |
| "loss": 1.2704, | |
| "step": 1395 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.45635080337524414, | |
| "learning_rate": 6.414825374198146e-07, | |
| "loss": 1.3303, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0249310731887817, | |
| "eval_runtime": 429.876, | |
| "eval_samples_per_second": 26.121, | |
| "eval_steps_per_second": 3.266, | |
| "step": 1400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 1403, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "total_flos": 5.282295580046131e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |