| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.9988002399520095, | |
| "eval_steps": 500, | |
| "global_step": 40000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07498500299940011, | |
| "grad_norm": 32.0, | |
| "learning_rate": 6.248437890527369e-06, | |
| "loss": 1.2095, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14997000599880023, | |
| "grad_norm": 16.5, | |
| "learning_rate": 1.2496875781054738e-05, | |
| "loss": 0.8665, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22495500899820037, | |
| "grad_norm": 17.75, | |
| "learning_rate": 1.8745313671582104e-05, | |
| "loss": 0.6844, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29994001199760045, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 2.4993751562109475e-05, | |
| "loss": 0.5099, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3749250149970006, | |
| "grad_norm": 7.375, | |
| "learning_rate": 3.124218945263684e-05, | |
| "loss": 0.2845, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.44991001799640074, | |
| "grad_norm": 8.0, | |
| "learning_rate": 3.749062734316421e-05, | |
| "loss": 0.2251, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5248950209958009, | |
| "grad_norm": 6.1875, | |
| "learning_rate": 4.373906523369158e-05, | |
| "loss": 0.206, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5998800239952009, | |
| "grad_norm": 3.640625, | |
| "learning_rate": 4.998750312421895e-05, | |
| "loss": 0.1874, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.674865026994601, | |
| "grad_norm": 4.8125, | |
| "learning_rate": 4.997630981947256e-05, | |
| "loss": 0.1793, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.7498500299940012, | |
| "grad_norm": 3.953125, | |
| "learning_rate": 4.9905094389741605e-05, | |
| "loss": 0.1708, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.8248350329934013, | |
| "grad_norm": 6.125, | |
| "learning_rate": 4.978648912557427e-05, | |
| "loss": 0.1614, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.8998200359928015, | |
| "grad_norm": 3.15625, | |
| "learning_rate": 4.962071971107133e-05, | |
| "loss": 0.1581, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9748050389922016, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 4.940810157507576e-05, | |
| "loss": 0.154, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6357488330240144, | |
| "eval_f1": 0.5235656092218528, | |
| "eval_loss": 0.1340431272983551, | |
| "eval_precision": 0.4623448805170602, | |
| "eval_recall": 0.6357488330240144, | |
| "eval_runtime": 40.157, | |
| "eval_samples_per_second": 295.191, | |
| "eval_steps_per_second": 18.453, | |
| "step": 6668 | |
| }, | |
| { | |
| "epoch": 1.0497900419916018, | |
| "grad_norm": 2.796875, | |
| "learning_rate": 4.914903929096945e-05, | |
| "loss": 0.1466, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.124775044991002, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 4.884402580684407e-05, | |
| "loss": 0.1503, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.1997600479904018, | |
| "grad_norm": 2.90625, | |
| "learning_rate": 4.8493641507511146e-05, | |
| "loss": 0.1463, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.274745050989802, | |
| "grad_norm": 5.15625, | |
| "learning_rate": 4.809855311013604e-05, | |
| "loss": 0.1473, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.349730053989202, | |
| "grad_norm": 3.3125, | |
| "learning_rate": 4.765951239559725e-05, | |
| "loss": 0.1448, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.4247150569886022, | |
| "grad_norm": 5.125, | |
| "learning_rate": 4.717735477798505e-05, | |
| "loss": 0.1381, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.4997000599880024, | |
| "grad_norm": 3.046875, | |
| "learning_rate": 4.665299771496145e-05, | |
| "loss": 0.1413, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.5746850629874025, | |
| "grad_norm": 4.40625, | |
| "learning_rate": 4.608743896200624e-05, | |
| "loss": 0.1419, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.6496700659868027, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 4.548175467387103e-05, | |
| "loss": 0.1407, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.7246550689862028, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 4.483709735685378e-05, | |
| "loss": 0.1394, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.799640071985603, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 4.415469367579033e-05, | |
| "loss": 0.1379, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.874625074985003, | |
| "grad_norm": 2.0, | |
| "learning_rate": 4.343584211993589e-05, | |
| "loss": 0.1377, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.9496100779844032, | |
| "grad_norm": 2.96875, | |
| "learning_rate": 4.268191053217765e-05, | |
| "loss": 0.1357, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6386662729880209, | |
| "eval_f1": 0.525822096019018, | |
| "eval_loss": 0.12524983286857605, | |
| "eval_precision": 0.463306913237047, | |
| "eval_recall": 0.6386662729880209, | |
| "eval_runtime": 40.1169, | |
| "eval_samples_per_second": 295.487, | |
| "eval_steps_per_second": 18.471, | |
| "step": 13336 | |
| }, | |
| { | |
| "epoch": 2.0245950809838034, | |
| "grad_norm": 5.4375, | |
| "learning_rate": 4.189433350628029e-05, | |
| "loss": 0.1366, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.0995800839832035, | |
| "grad_norm": 5.875, | |
| "learning_rate": 4.10746096571167e-05, | |
| "loss": 0.1329, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.1745650869826036, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 4.02242987690783e-05, | |
| "loss": 0.1325, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.249550089982004, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 3.9345018828090864e-05, | |
| "loss": 0.1321, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.324535092981404, | |
| "grad_norm": 2.796875, | |
| "learning_rate": 3.843844294288368e-05, | |
| "loss": 0.1332, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.3995200959808036, | |
| "grad_norm": 3.203125, | |
| "learning_rate": 3.750629616136989e-05, | |
| "loss": 0.1281, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.4745050989802038, | |
| "grad_norm": 2.921875, | |
| "learning_rate": 3.6550352188196244e-05, | |
| "loss": 0.1307, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.549490101979604, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 3.557243000970787e-05, | |
| "loss": 0.1345, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.624475104979004, | |
| "grad_norm": 3.28125, | |
| "learning_rate": 3.457439043275033e-05, | |
| "loss": 0.129, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.699460107978404, | |
| "grad_norm": 3.703125, | |
| "learning_rate": 3.355813254389495e-05, | |
| "loss": 0.135, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.7744451109778043, | |
| "grad_norm": 2.96875, | |
| "learning_rate": 3.252559009582478e-05, | |
| "loss": 0.1336, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.8494301139772045, | |
| "grad_norm": 3.6875, | |
| "learning_rate": 3.14787278277573e-05, | |
| "loss": 0.1327, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.9244151169766046, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 3.0419537726905434e-05, | |
| "loss": 0.1326, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.9994001199760048, | |
| "grad_norm": 4.28125, | |
| "learning_rate": 2.9350035238090666e-05, | |
| "loss": 0.1295, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6393481806422586, | |
| "eval_f1": 0.5262540403072158, | |
| "eval_loss": 0.12466703355312347, | |
| "eval_precision": 0.4630570055607114, | |
| "eval_recall": 0.6393481806422586, | |
| "eval_runtime": 40.1133, | |
| "eval_samples_per_second": 295.513, | |
| "eval_steps_per_second": 18.473, | |
| "step": 20004 | |
| }, | |
| { | |
| "epoch": 3.074385122975405, | |
| "grad_norm": 4.625, | |
| "learning_rate": 2.827225542872053e-05, | |
| "loss": 0.1301, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.149370125974805, | |
| "grad_norm": 3.4375, | |
| "learning_rate": 2.7188249116427988e-05, | |
| "loss": 0.1287, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.224355128974205, | |
| "grad_norm": 4.0625, | |
| "learning_rate": 2.6100078966740953e-05, | |
| "loss": 0.1314, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.2993401319736053, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 2.500981556820753e-05, | |
| "loss": 0.1299, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.3743251349730055, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 2.3919533492445064e-05, | |
| "loss": 0.1298, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.4493101379724056, | |
| "grad_norm": 4.25, | |
| "learning_rate": 2.2831307346610255e-05, | |
| "loss": 0.1293, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.5242951409718057, | |
| "grad_norm": 3.640625, | |
| "learning_rate": 2.17472078258016e-05, | |
| "loss": 0.1261, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.599280143971206, | |
| "grad_norm": 3.75, | |
| "learning_rate": 2.066929777290578e-05, | |
| "loss": 0.1298, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.674265146970606, | |
| "grad_norm": 4.9375, | |
| "learning_rate": 1.9599628253385327e-05, | |
| "loss": 0.1302, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.749250149970006, | |
| "grad_norm": 3.4375, | |
| "learning_rate": 1.8540234652476617e-05, | |
| "loss": 0.1295, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.824235152969406, | |
| "grad_norm": 3.875, | |
| "learning_rate": 1.7493132802224482e-05, | |
| "loss": 0.1285, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.8992201559688064, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 1.6460315145722894e-05, | |
| "loss": 0.1272, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.974205158968206, | |
| "grad_norm": 3.890625, | |
| "learning_rate": 1.5443746945860566e-05, | |
| "loss": 0.1294, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.639137281367752, | |
| "eval_f1": 0.5262984000216192, | |
| "eval_loss": 0.12251855432987213, | |
| "eval_precision": 0.4640513156681777, | |
| "eval_recall": 0.639137281367752, | |
| "eval_runtime": 40.1904, | |
| "eval_samples_per_second": 294.946, | |
| "eval_steps_per_second": 18.437, | |
| "step": 26672 | |
| }, | |
| { | |
| "epoch": 4.049190161967607, | |
| "grad_norm": 3.1875, | |
| "learning_rate": 1.4445362545785581e-05, | |
| "loss": 0.1311, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.124175164967006, | |
| "grad_norm": 3.390625, | |
| "learning_rate": 1.3467061688204524e-05, | |
| "loss": 0.1289, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.199160167966407, | |
| "grad_norm": 6.0625, | |
| "learning_rate": 1.2510705900519926e-05, | |
| "loss": 0.1263, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.274145170965807, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 1.1578114952684529e-05, | |
| "loss": 0.1273, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 4.349130173965207, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 1.0671063394512279e-05, | |
| "loss": 0.1339, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.424115176964607, | |
| "grad_norm": 4.5, | |
| "learning_rate": 9.791277179034853e-06, | |
| "loss": 0.1286, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.499100179964008, | |
| "grad_norm": 3.921875, | |
| "learning_rate": 8.940430378329174e-06, | |
| "loss": 0.133, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.574085182963407, | |
| "grad_norm": 2.828125, | |
| "learning_rate": 8.120141998064757e-06, | |
| "loss": 0.1241, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.649070185962808, | |
| "grad_norm": 3.203125, | |
| "learning_rate": 7.331972896832292e-06, | |
| "loss": 0.1269, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.7240551889622076, | |
| "grad_norm": 6.40625, | |
| "learning_rate": 6.577422816115633e-06, | |
| "loss": 0.1303, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.799040191961607, | |
| "grad_norm": 3.328125, | |
| "learning_rate": 5.857927526558302e-06, | |
| "loss": 0.1306, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.874025194961008, | |
| "grad_norm": 2.96875, | |
| "learning_rate": 5.174856095954883e-06, | |
| "loss": 0.129, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.9490101979604075, | |
| "grad_norm": 2.65625, | |
| "learning_rate": 4.529508284165662e-06, | |
| "loss": 0.1271, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6391864911984703, | |
| "eval_f1": 0.5263388946843862, | |
| "eval_loss": 0.12287881225347519, | |
| "eval_precision": 0.464080708996073, | |
| "eval_recall": 0.6391864911984703, | |
| "eval_runtime": 40.1821, | |
| "eval_samples_per_second": 295.007, | |
| "eval_steps_per_second": 18.441, | |
| "step": 33340 | |
| }, | |
| { | |
| "epoch": 5.023995200959808, | |
| "grad_norm": 2.296875, | |
| "learning_rate": 3.923112069911616e-06, | |
| "loss": 0.1263, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 5.098980203959208, | |
| "grad_norm": 8.375, | |
| "learning_rate": 3.3568213141557586e-06, | |
| "loss": 0.1289, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 5.173965206958608, | |
| "grad_norm": 3.21875, | |
| "learning_rate": 2.8317135645169912e-06, | |
| "loss": 0.1291, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 5.248950209958008, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 2.3487880048942728e-06, | |
| "loss": 0.1271, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 5.323935212957409, | |
| "grad_norm": 3.765625, | |
| "learning_rate": 1.9089635542026586e-06, | |
| "loss": 0.1306, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 5.398920215956808, | |
| "grad_norm": 6.21875, | |
| "learning_rate": 1.5130771178388153e-06, | |
| "loss": 0.1287, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 5.473905218956209, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 1.1618819952033066e-06, | |
| "loss": 0.1268, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 5.548890221955609, | |
| "grad_norm": 3.625, | |
| "learning_rate": 8.560464463097773e-07, | |
| "loss": 0.1293, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 5.623875224955009, | |
| "grad_norm": 4.28125, | |
| "learning_rate": 5.961524202084901e-07, | |
| "loss": 0.1292, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 5.698860227954409, | |
| "grad_norm": 3.03125, | |
| "learning_rate": 3.826944476438388e-07, | |
| "loss": 0.1311, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.7738452309538095, | |
| "grad_norm": 3.609375, | |
| "learning_rate": 2.1607870005291575e-07, | |
| "loss": 0.1269, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.848830233953209, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 9.662221669560767e-08, | |
| "loss": 0.1273, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.92381523695261, | |
| "grad_norm": 3.15625, | |
| "learning_rate": 2.4552301386951415e-08, | |
| "loss": 0.1312, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 5.9988002399520095, | |
| "grad_norm": 4.78125, | |
| "learning_rate": 6.089979592838191e-12, | |
| "loss": 0.1285, | |
| "step": 40000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 40008, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 1000, | |
| "total_flos": 1.7186915377741824e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |