| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7152436298614215, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001788109074653554, | |
| "eval_loss": 2.046928882598877, | |
| "eval_runtime": 19.3171, | |
| "eval_samples_per_second": 12.217, | |
| "eval_steps_per_second": 6.109, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008940545373267769, | |
| "grad_norm": 12.01391887664795, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 7.1881, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.017881090746535537, | |
| "grad_norm": 16.71553611755371, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 6.7995, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02682163611980331, | |
| "grad_norm": 31.485000610351562, | |
| "learning_rate": 5e-05, | |
| "loss": 4.923, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.035762181493071074, | |
| "grad_norm": 23.379528045654297, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 5.7787, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.044702726866338846, | |
| "grad_norm": 18.306440353393555, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 4.6506, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05364327223960662, | |
| "grad_norm": 25.282123565673828, | |
| "learning_rate": 0.0001, | |
| "loss": 4.8731, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06258381761287439, | |
| "grad_norm": 22.050247192382812, | |
| "learning_rate": 9.995494831023409e-05, | |
| "loss": 4.7905, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.07152436298614215, | |
| "grad_norm": 36.01411056518555, | |
| "learning_rate": 9.981987442712633e-05, | |
| "loss": 6.3107, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08046490835940992, | |
| "grad_norm": 41.16471481323242, | |
| "learning_rate": 9.959502176294383e-05, | |
| "loss": 5.1572, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08940545373267769, | |
| "grad_norm": 50.630348205566406, | |
| "learning_rate": 9.928079551738543e-05, | |
| "loss": 5.9515, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09834599910594546, | |
| "grad_norm": 10.866789817810059, | |
| "learning_rate": 9.887776194738432e-05, | |
| "loss": 6.7468, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.10728654447921324, | |
| "grad_norm": 11.329707145690918, | |
| "learning_rate": 9.838664734667495e-05, | |
| "loss": 6.4092, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11622708985248101, | |
| "grad_norm": 12.150516510009766, | |
| "learning_rate": 9.780833673696254e-05, | |
| "loss": 4.6159, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.12516763522574878, | |
| "grad_norm": 12.043376922607422, | |
| "learning_rate": 9.714387227305422e-05, | |
| "loss": 4.6607, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13410818059901655, | |
| "grad_norm": 13.809493064880371, | |
| "learning_rate": 9.639445136482548e-05, | |
| "loss": 4.6828, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1430487259722843, | |
| "grad_norm": 17.044633865356445, | |
| "learning_rate": 9.55614245194068e-05, | |
| "loss": 4.1347, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15198927134555207, | |
| "grad_norm": 17.997037887573242, | |
| "learning_rate": 9.464629290747842e-05, | |
| "loss": 4.9318, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.16092981671881984, | |
| "grad_norm": 29.060178756713867, | |
| "learning_rate": 9.365070565805941e-05, | |
| "loss": 5.6566, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1698703620920876, | |
| "grad_norm": 37.676151275634766, | |
| "learning_rate": 9.257645688666556e-05, | |
| "loss": 4.9255, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.17881090746535538, | |
| "grad_norm": 40.094573974609375, | |
| "learning_rate": 9.142548246219212e-05, | |
| "loss": 5.421, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17881090746535538, | |
| "eval_loss": 1.4111511707305908, | |
| "eval_runtime": 19.669, | |
| "eval_samples_per_second": 11.999, | |
| "eval_steps_per_second": 5.999, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18775145283862316, | |
| "grad_norm": 8.953372955322266, | |
| "learning_rate": 9.019985651834703e-05, | |
| "loss": 6.6067, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.19669199821189093, | |
| "grad_norm": 9.89356803894043, | |
| "learning_rate": 8.890178771592199e-05, | |
| "loss": 5.5489, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2056325435851587, | |
| "grad_norm": 10.751019477844238, | |
| "learning_rate": 8.753361526263621e-05, | |
| "loss": 4.145, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.21457308895842647, | |
| "grad_norm": 11.124737739562988, | |
| "learning_rate": 8.609780469772623e-05, | |
| "loss": 3.9116, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.22351363433169424, | |
| "grad_norm": 13.146486282348633, | |
| "learning_rate": 8.459694344887732e-05, | |
| "loss": 4.5305, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.23245417970496202, | |
| "grad_norm": 20.332517623901367, | |
| "learning_rate": 8.303373616950408e-05, | |
| "loss": 4.6298, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.24139472507822976, | |
| "grad_norm": 15.850321769714355, | |
| "learning_rate": 8.141099986478212e-05, | |
| "loss": 5.1285, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.25033527045149756, | |
| "grad_norm": 25.98038673400879, | |
| "learning_rate": 7.973165881521434e-05, | |
| "loss": 4.8565, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25927581582476533, | |
| "grad_norm": 41.79204559326172, | |
| "learning_rate": 7.799873930687978e-05, | |
| "loss": 5.2209, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.2682163611980331, | |
| "grad_norm": 66.45893096923828, | |
| "learning_rate": 7.621536417786159e-05, | |
| "loss": 5.4458, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2771569065713009, | |
| "grad_norm": 7.945077896118164, | |
| "learning_rate": 7.438474719068173e-05, | |
| "loss": 6.6603, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.2860974519445686, | |
| "grad_norm": 11.32933521270752, | |
| "learning_rate": 7.251018724088367e-05, | |
| "loss": 5.9139, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.29503799731783636, | |
| "grad_norm": 9.906847953796387, | |
| "learning_rate": 7.059506241219965e-05, | |
| "loss": 5.6972, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.30397854269110414, | |
| "grad_norm": 11.90020751953125, | |
| "learning_rate": 6.864282388901544e-05, | |
| "loss": 4.1571, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3129190880643719, | |
| "grad_norm": 11.035331726074219, | |
| "learning_rate": 6.665698973710288e-05, | |
| "loss": 4.3225, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3218596334376397, | |
| "grad_norm": 16.212894439697266, | |
| "learning_rate": 6.464113856382752e-05, | |
| "loss": 4.0366, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.33080017881090745, | |
| "grad_norm": 14.783008575439453, | |
| "learning_rate": 6.259890306925627e-05, | |
| "loss": 3.9975, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.3397407241841752, | |
| "grad_norm": 20.191490173339844, | |
| "learning_rate": 6.0533963499786314e-05, | |
| "loss": 4.8397, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.348681269557443, | |
| "grad_norm": 19.51316261291504, | |
| "learning_rate": 5.8450041016092464e-05, | |
| "loss": 3.8329, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.35762181493071077, | |
| "grad_norm": 97.10255432128906, | |
| "learning_rate": 5.6350890987343944e-05, | |
| "loss": 6.9008, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35762181493071077, | |
| "eval_loss": 1.2790147066116333, | |
| "eval_runtime": 19.6612, | |
| "eval_samples_per_second": 12.003, | |
| "eval_steps_per_second": 6.002, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36656236030397854, | |
| "grad_norm": 9.973484992980957, | |
| "learning_rate": 5.4240296223775465e-05, | |
| "loss": 6.283, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.3755029056772463, | |
| "grad_norm": 9.57100772857666, | |
| "learning_rate": 5.212206015980742e-05, | |
| "loss": 5.2025, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3844434510505141, | |
| "grad_norm": 11.266950607299805, | |
| "learning_rate": 5e-05, | |
| "loss": 4.9933, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.39338399642378186, | |
| "grad_norm": 15.667479515075684, | |
| "learning_rate": 4.78779398401926e-05, | |
| "loss": 4.0976, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.40232454179704963, | |
| "grad_norm": 14.551920890808105, | |
| "learning_rate": 4.575970377622456e-05, | |
| "loss": 3.9731, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4112650871703174, | |
| "grad_norm": 9.969139099121094, | |
| "learning_rate": 4.364910901265606e-05, | |
| "loss": 4.0724, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4202056325435852, | |
| "grad_norm": 13.29488754272461, | |
| "learning_rate": 4.1549958983907555e-05, | |
| "loss": 4.0002, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.42914617791685294, | |
| "grad_norm": 17.779531478881836, | |
| "learning_rate": 3.94660365002137e-05, | |
| "loss": 4.5389, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4380867232901207, | |
| "grad_norm": 39.08097839355469, | |
| "learning_rate": 3.740109693074375e-05, | |
| "loss": 4.8595, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.4470272686633885, | |
| "grad_norm": 51.12818908691406, | |
| "learning_rate": 3.5358861436172485e-05, | |
| "loss": 5.6399, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.45596781403665626, | |
| "grad_norm": 8.17031192779541, | |
| "learning_rate": 3.334301026289712e-05, | |
| "loss": 6.488, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.46490835940992403, | |
| "grad_norm": 11.224337577819824, | |
| "learning_rate": 3.135717611098458e-05, | |
| "loss": 4.4233, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.47384890478319175, | |
| "grad_norm": 11.139561653137207, | |
| "learning_rate": 2.9404937587800375e-05, | |
| "loss": 4.3712, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.4827894501564595, | |
| "grad_norm": 9.91527271270752, | |
| "learning_rate": 2.748981275911633e-05, | |
| "loss": 4.235, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4917299955297273, | |
| "grad_norm": 11.947907447814941, | |
| "learning_rate": 2.5615252809318284e-05, | |
| "loss": 3.5619, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5006705409029951, | |
| "grad_norm": 14.57845687866211, | |
| "learning_rate": 2.3784635822138424e-05, | |
| "loss": 4.2477, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5096110862762628, | |
| "grad_norm": 16.487865447998047, | |
| "learning_rate": 2.2001260693120233e-05, | |
| "loss": 4.7886, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.5185516316495307, | |
| "grad_norm": 25.24747657775879, | |
| "learning_rate": 2.026834118478567e-05, | |
| "loss": 3.643, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5274921770227984, | |
| "grad_norm": 29.675588607788086, | |
| "learning_rate": 1.858900013521788e-05, | |
| "loss": 4.3066, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.5364327223960662, | |
| "grad_norm": 42.390953063964844, | |
| "learning_rate": 1.6966263830495936e-05, | |
| "loss": 4.9912, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5364327223960662, | |
| "eval_loss": 1.2248698472976685, | |
| "eval_runtime": 19.6654, | |
| "eval_samples_per_second": 12.001, | |
| "eval_steps_per_second": 6.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5453732677693339, | |
| "grad_norm": 6.869641304016113, | |
| "learning_rate": 1.5403056551122697e-05, | |
| "loss": 5.6695, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.5543138131426018, | |
| "grad_norm": 12.490229606628418, | |
| "learning_rate": 1.3902195302273779e-05, | |
| "loss": 4.9411, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5632543585158695, | |
| "grad_norm": 12.478997230529785, | |
| "learning_rate": 1.246638473736378e-05, | |
| "loss": 4.2023, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.5721949038891372, | |
| "grad_norm": 12.402631759643555, | |
| "learning_rate": 1.1098212284078036e-05, | |
| "loss": 4.0082, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.581135449262405, | |
| "grad_norm": 12.362221717834473, | |
| "learning_rate": 9.800143481652979e-06, | |
| "loss": 4.2276, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5900759946356727, | |
| "grad_norm": 12.496500015258789, | |
| "learning_rate": 8.574517537807897e-06, | |
| "loss": 3.4944, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5990165400089406, | |
| "grad_norm": 25.300518035888672, | |
| "learning_rate": 7.423543113334436e-06, | |
| "loss": 4.4454, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.6079570853822083, | |
| "grad_norm": 19.98192596435547, | |
| "learning_rate": 6.349294341940593e-06, | |
| "loss": 4.4782, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6168976307554761, | |
| "grad_norm": 34.964942932128906, | |
| "learning_rate": 5.353707092521582e-06, | |
| "loss": 3.9177, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.6258381761287438, | |
| "grad_norm": 37.618648529052734, | |
| "learning_rate": 4.43857548059321e-06, | |
| "loss": 5.8998, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6347787215020116, | |
| "grad_norm": 12.334329605102539, | |
| "learning_rate": 3.605548635174533e-06, | |
| "loss": 5.8214, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.6437192668752794, | |
| "grad_norm": 11.229202270507812, | |
| "learning_rate": 2.85612772694579e-06, | |
| "loss": 4.0393, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6526598122485472, | |
| "grad_norm": 10.08092212677002, | |
| "learning_rate": 2.191663263037458e-06, | |
| "loss": 4.3505, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.6616003576218149, | |
| "grad_norm": 9.676584243774414, | |
| "learning_rate": 1.6133526533250565e-06, | |
| "loss": 3.7247, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6705409029950827, | |
| "grad_norm": 10.292547225952148, | |
| "learning_rate": 1.1222380526156928e-06, | |
| "loss": 3.5345, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.6794814483683504, | |
| "grad_norm": 15.131571769714355, | |
| "learning_rate": 7.192044826145771e-07, | |
| "loss": 3.9341, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6884219937416183, | |
| "grad_norm": 19.556400299072266, | |
| "learning_rate": 4.049782370561583e-07, | |
| "loss": 4.0314, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.697362539114886, | |
| "grad_norm": 22.79786491394043, | |
| "learning_rate": 1.8012557287367392e-07, | |
| "loss": 4.756, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7063030844881538, | |
| "grad_norm": 29.830158233642578, | |
| "learning_rate": 4.5051689765929214e-08, | |
| "loss": 4.7687, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.7152436298614215, | |
| "grad_norm": 40.145057678222656, | |
| "learning_rate": 0.0, | |
| "loss": 4.1833, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7152436298614215, | |
| "eval_loss": 1.181886076927185, | |
| "eval_runtime": 19.6499, | |
| "eval_samples_per_second": 12.01, | |
| "eval_steps_per_second": 6.005, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4091489359010202e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |