| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.984771573604061, | |
| "eval_steps": 50, | |
| "global_step": 441, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0338409475465313, | |
| "grad_norm": 19.362921643157705, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7543, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0676818950930626, | |
| "grad_norm": 12.440471039796169, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5921, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10152284263959391, | |
| "grad_norm": 7.151840796848598, | |
| "learning_rate": 9.996679701338661e-07, | |
| "loss": 1.2895, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1353637901861252, | |
| "grad_norm": 4.254605069904108, | |
| "learning_rate": 9.986723215107924e-07, | |
| "loss": 1.135, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1692047377326565, | |
| "grad_norm": 3.6917219941972412, | |
| "learning_rate": 9.97014376471095e-07, | |
| "loss": 1.0651, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 3.7987347848198634, | |
| "learning_rate": 9.946963369638524e-07, | |
| "loss": 1.0038, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23688663282571912, | |
| "grad_norm": 3.637845138860867, | |
| "learning_rate": 9.917212816224536e-07, | |
| "loss": 0.9911, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2707275803722504, | |
| "grad_norm": 3.3460567231716873, | |
| "learning_rate": 9.880931616758056e-07, | |
| "loss": 0.9617, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.30456852791878175, | |
| "grad_norm": 3.377056006196627, | |
| "learning_rate": 9.838167957006293e-07, | |
| "loss": 0.955, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "grad_norm": 3.3982403942088624, | |
| "learning_rate": 9.788978632218138e-07, | |
| "loss": 0.9458, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.338409475465313, | |
| "eval_loss": 0.9426366686820984, | |
| "eval_runtime": 73.0561, | |
| "eval_samples_per_second": 57.49, | |
| "eval_steps_per_second": 0.903, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.37225042301184436, | |
| "grad_norm": 3.689104664245426, | |
| "learning_rate": 9.73342897169329e-07, | |
| "loss": 0.943, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 3.395123607137787, | |
| "learning_rate": 9.671592752017137e-07, | |
| "loss": 0.9093, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.43993231810490696, | |
| "grad_norm": 3.357446602126202, | |
| "learning_rate": 9.603552099076648e-07, | |
| "loss": 0.9195, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.47377326565143824, | |
| "grad_norm": 3.691462819097496, | |
| "learning_rate": 9.52939737898737e-07, | |
| "loss": 0.886, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 3.5969341059847415, | |
| "learning_rate": 9.449227078076443e-07, | |
| "loss": 0.9185, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5414551607445008, | |
| "grad_norm": 3.558555135432629, | |
| "learning_rate": 9.363147672080985e-07, | |
| "loss": 0.8983, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5752961082910322, | |
| "grad_norm": 3.6252630464826523, | |
| "learning_rate": 9.271273484735592e-07, | |
| "loss": 0.8843, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 3.3121290866804913, | |
| "learning_rate": 9.173726535936766e-07, | |
| "loss": 0.8844, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6429780033840947, | |
| "grad_norm": 3.580257416417439, | |
| "learning_rate": 9.070636379685915e-07, | |
| "loss": 0.8994, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "grad_norm": 3.629629801061202, | |
| "learning_rate": 8.962139932026156e-07, | |
| "loss": 0.8779, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.676818950930626, | |
| "eval_loss": 0.887722373008728, | |
| "eval_runtime": 72.6917, | |
| "eval_samples_per_second": 57.778, | |
| "eval_steps_per_second": 0.908, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7106598984771574, | |
| "grad_norm": 3.3980001711357466, | |
| "learning_rate": 8.848381289201459e-07, | |
| "loss": 0.864, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.7445008460236887, | |
| "grad_norm": 3.37541287934786, | |
| "learning_rate": 8.72951153627962e-07, | |
| "loss": 0.8795, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7783417935702199, | |
| "grad_norm": 5.903555404027213, | |
| "learning_rate": 8.605688546493238e-07, | |
| "loss": 0.8657, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 3.3534687236406615, | |
| "learning_rate": 8.477076771565202e-07, | |
| "loss": 0.8627, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8460236886632826, | |
| "grad_norm": 3.472402321033115, | |
| "learning_rate": 8.343847023297169e-07, | |
| "loss": 0.8695, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8798646362098139, | |
| "grad_norm": 3.5485673752671865, | |
| "learning_rate": 8.206176246711065e-07, | |
| "loss": 0.8429, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9137055837563451, | |
| "grad_norm": 3.4858020678381054, | |
| "learning_rate": 8.064247285044972e-07, | |
| "loss": 0.8603, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.9475465313028765, | |
| "grad_norm": 3.7500528392849324, | |
| "learning_rate": 7.918248636915459e-07, | |
| "loss": 0.8622, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9813874788494078, | |
| "grad_norm": 3.409665848861188, | |
| "learning_rate": 7.768374205968906e-07, | |
| "loss": 0.8473, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "grad_norm": 3.6326130834418944, | |
| "learning_rate": 7.614823043354285e-07, | |
| "loss": 0.82, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.015228426395939, | |
| "eval_loss": 0.8625058531761169, | |
| "eval_runtime": 72.7377, | |
| "eval_samples_per_second": 57.742, | |
| "eval_steps_per_second": 0.907, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0490693739424704, | |
| "grad_norm": 3.6224664863477494, | |
| "learning_rate": 7.457799083359471e-07, | |
| "loss": 0.823, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.0829103214890017, | |
| "grad_norm": 3.392243863011647, | |
| "learning_rate": 7.297510872562131e-07, | |
| "loss": 0.7903, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.116751269035533, | |
| "grad_norm": 3.6101258998571875, | |
| "learning_rate": 7.134171292854955e-07, | |
| "loss": 0.7844, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.1505922165820643, | |
| "grad_norm": 3.4039338742997685, | |
| "learning_rate": 6.967997278713089e-07, | |
| "loss": 0.7888, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1844331641285957, | |
| "grad_norm": 3.3872407965058384, | |
| "learning_rate": 6.79920952907921e-07, | |
| "loss": 0.7968, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.218274111675127, | |
| "grad_norm": 3.5554493985038684, | |
| "learning_rate": 6.628032214248982e-07, | |
| "loss": 0.7823, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.252115059221658, | |
| "grad_norm": 3.655696778331916, | |
| "learning_rate": 6.454692678146119e-07, | |
| "loss": 0.7848, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.2859560067681894, | |
| "grad_norm": 3.459528935552729, | |
| "learning_rate": 6.279421136382494e-07, | |
| "loss": 0.7683, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3197969543147208, | |
| "grad_norm": 3.506826792292506, | |
| "learning_rate": 6.102450370504299e-07, | |
| "loss": 0.7776, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "grad_norm": 3.53257152623125, | |
| "learning_rate": 5.924015418830354e-07, | |
| "loss": 0.7763, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.353637901861252, | |
| "eval_loss": 0.8517683148384094, | |
| "eval_runtime": 72.7613, | |
| "eval_samples_per_second": 57.723, | |
| "eval_steps_per_second": 0.907, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3874788494077834, | |
| "grad_norm": 3.4808019080846138, | |
| "learning_rate": 5.74435326429313e-07, | |
| "loss": 0.7586, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.4213197969543148, | |
| "grad_norm": 3.4920276219604696, | |
| "learning_rate": 5.563702519697108e-07, | |
| "loss": 0.7693, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.455160744500846, | |
| "grad_norm": 3.563387899492764, | |
| "learning_rate": 5.382303110812466e-07, | |
| "loss": 0.7784, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.4890016920473772, | |
| "grad_norm": 3.3319886045415914, | |
| "learning_rate": 5.200395957725005e-07, | |
| "loss": 0.7789, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.5228426395939088, | |
| "grad_norm": 3.824253256636034, | |
| "learning_rate": 5.018222654865471e-07, | |
| "loss": 0.7692, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5566835871404399, | |
| "grad_norm": 3.49580497047461, | |
| "learning_rate": 4.836025150143318e-07, | |
| "loss": 0.7954, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5905245346869712, | |
| "grad_norm": 3.726391320444098, | |
| "learning_rate": 4.654045423610952e-07, | |
| "loss": 0.7882, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.6243654822335025, | |
| "grad_norm": 3.940367355899551, | |
| "learning_rate": 4.4725251660853357e-07, | |
| "loss": 0.7782, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6582064297800339, | |
| "grad_norm": 3.603115659587366, | |
| "learning_rate": 4.2917054581536926e-07, | |
| "loss": 0.7709, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "grad_norm": 3.5695746889936726, | |
| "learning_rate": 4.1118264499897003e-07, | |
| "loss": 0.7719, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6920473773265652, | |
| "eval_loss": 0.8421301245689392, | |
| "eval_runtime": 72.7217, | |
| "eval_samples_per_second": 57.754, | |
| "eval_steps_per_second": 0.908, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7258883248730963, | |
| "grad_norm": 3.654464869264395, | |
| "learning_rate": 3.9331270424053616e-07, | |
| "loss": 0.7599, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.7597292724196278, | |
| "grad_norm": 3.669706546573186, | |
| "learning_rate": 3.755844569562191e-07, | |
| "loss": 0.7727, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.793570219966159, | |
| "grad_norm": 3.514197357489897, | |
| "learning_rate": 3.580214483763093e-07, | |
| "loss": 0.7709, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.8274111675126905, | |
| "grad_norm": 3.8443336447359244, | |
| "learning_rate": 3.406470042743574e-07, | |
| "loss": 0.782, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8612521150592216, | |
| "grad_norm": 3.625482123749312, | |
| "learning_rate": 3.23484199987761e-07, | |
| "loss": 0.7793, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.895093062605753, | |
| "grad_norm": 3.5197175718189317, | |
| "learning_rate": 3.065558297709588e-07, | |
| "loss": 0.7623, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9289340101522843, | |
| "grad_norm": 3.4330767306351935, | |
| "learning_rate": 2.898843765219388e-07, | |
| "loss": 0.7628, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.9627749576988156, | |
| "grad_norm": 3.822061440419418, | |
| "learning_rate": 2.7349198192226295e-07, | |
| "loss": 0.7584, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.996615905245347, | |
| "grad_norm": 3.553974834774039, | |
| "learning_rate": 2.574004170302696e-07, | |
| "loss": 0.7645, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "grad_norm": 3.9479223535228574, | |
| "learning_rate": 2.4163105336650643e-07, | |
| "loss": 0.7347, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.030456852791878, | |
| "eval_loss": 0.8380723595619202, | |
| "eval_runtime": 72.7338, | |
| "eval_samples_per_second": 57.745, | |
| "eval_steps_per_second": 0.907, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0642978003384096, | |
| "grad_norm": 3.7132773157030425, | |
| "learning_rate": 2.2620483452979887e-07, | |
| "loss": 0.7348, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.0981387478849407, | |
| "grad_norm": 3.6057007836329147, | |
| "learning_rate": 2.1114224838164806e-07, | |
| "loss": 0.7193, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1319796954314723, | |
| "grad_norm": 3.6945593121525286, | |
| "learning_rate": 1.964632998359036e-07, | |
| "loss": 0.7286, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.1658206429780034, | |
| "grad_norm": 3.9031426698674663, | |
| "learning_rate": 1.8218748428984782e-07, | |
| "loss": 0.7024, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.199661590524535, | |
| "grad_norm": 3.4579222711663786, | |
| "learning_rate": 1.6833376173198005e-07, | |
| "loss": 0.7084, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.233502538071066, | |
| "grad_norm": 3.719715731546776, | |
| "learning_rate": 1.5492053156088498e-07, | |
| "loss": 0.7088, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.267343485617597, | |
| "grad_norm": 3.8269194060342446, | |
| "learning_rate": 1.4196560814863567e-07, | |
| "loss": 0.7244, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.3011844331641287, | |
| "grad_norm": 3.839398221167279, | |
| "learning_rate": 1.294861971811773e-07, | |
| "loss": 0.7261, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.33502538071066, | |
| "grad_norm": 3.6636890535123885, | |
| "learning_rate": 1.1749887280712161e-07, | |
| "loss": 0.7193, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "grad_norm": 3.9676568766055165, | |
| "learning_rate": 1.0601955562529774e-07, | |
| "loss": 0.7232, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3688663282571913, | |
| "eval_loss": 0.84018474817276, | |
| "eval_runtime": 72.7538, | |
| "eval_samples_per_second": 57.729, | |
| "eval_steps_per_second": 0.907, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4027072758037225, | |
| "grad_norm": 3.871757730003177, | |
| "learning_rate": 9.506349154029425e-08, | |
| "loss": 0.7072, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.436548223350254, | |
| "grad_norm": 3.709431977886577, | |
| "learning_rate": 8.46452315140772e-08, | |
| "loss": 0.724, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.470389170896785, | |
| "grad_norm": 3.778002504886439, | |
| "learning_rate": 7.477861224057403e-08, | |
| "loss": 0.726, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.504230118443316, | |
| "grad_norm": 4.034569033748593, | |
| "learning_rate": 6.547673776889095e-08, | |
| "loss": 0.7152, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5380710659898478, | |
| "grad_norm": 3.6052144662756582, | |
| "learning_rate": 5.6751962099570396e-08, | |
| "loss": 0.7222, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.571912013536379, | |
| "grad_norm": 3.6412688810760887, | |
| "learning_rate": 4.861587277700274e-08, | |
| "loss": 0.7141, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.6057529610829104, | |
| "grad_norm": 3.788492541312578, | |
| "learning_rate": 4.107927549978235e-08, | |
| "loss": 0.7075, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.6395939086294415, | |
| "grad_norm": 3.7740858241104385, | |
| "learning_rate": 3.4152179769449396e-08, | |
| "loss": 0.7171, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.673434856175973, | |
| "grad_norm": 3.693130131986868, | |
| "learning_rate": 2.784378559667622e-08, | |
| "loss": 0.7198, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "grad_norm": 3.627471625901726, | |
| "learning_rate": 2.2162471282553553e-08, | |
| "loss": 0.721, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.707275803722504, | |
| "eval_loss": 0.8391405940055847, | |
| "eval_runtime": 72.6846, | |
| "eval_samples_per_second": 57.784, | |
| "eval_steps_per_second": 0.908, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7411167512690353, | |
| "grad_norm": 3.7944346677217715, | |
| "learning_rate": 1.7115782291206082e-08, | |
| "loss": 0.7091, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.774957698815567, | |
| "grad_norm": 3.684276269607114, | |
| "learning_rate": 1.2710421228514733e-08, | |
| "loss": 0.7247, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.808798646362098, | |
| "grad_norm": 3.551925911903825, | |
| "learning_rate": 8.952238940255153e-09, | |
| "loss": 0.7245, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.8426395939086295, | |
| "grad_norm": 3.549163935230389, | |
| "learning_rate": 5.846226741475557e-09, | |
| "loss": 0.7254, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.8764805414551606, | |
| "grad_norm": 3.809089531185781, | |
| "learning_rate": 3.3965097874343872e-09, | |
| "loss": 0.7037, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.910321489001692, | |
| "grad_norm": 3.5360521655033934, | |
| "learning_rate": 1.6063415949008618e-09, | |
| "loss": 0.7319, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.9441624365482233, | |
| "grad_norm": 3.8076879389100875, | |
| "learning_rate": 4.780997210962478e-10, | |
| "loss": 0.7199, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.9780033840947544, | |
| "grad_norm": 3.6857618291895236, | |
| "learning_rate": 1.328260601385356e-11, | |
| "loss": 0.7036, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.984771573604061, | |
| "step": 441, | |
| "total_flos": 2600101931384832.0, | |
| "train_loss": 0.8256486911081673, | |
| "train_runtime": 6391.2706, | |
| "train_samples_per_second": 17.743, | |
| "train_steps_per_second": 0.069 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 441, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2600101931384832.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |