| { |
| "best_global_step": 456, |
| "best_metric": 0.008637478575110435, |
| "best_model_checkpoint": "saves_multiple/lora/llama-3-8b-instruct/train_cb_101112_1760637983/checkpoint-456", |
| "epoch": 20.0, |
| "eval_steps": 57, |
| "global_step": 1140, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.08771929824561403, |
| "grad_norm": 7.695529937744141, |
| "learning_rate": 1.7543859649122807e-06, |
| "loss": 0.9474, |
| "num_input_tokens_seen": 3648, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 8.015158653259277, |
| "learning_rate": 3.9473684210526315e-06, |
| "loss": 1.1547, |
| "num_input_tokens_seen": 6848, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 7.584747314453125, |
| "learning_rate": 6.140350877192982e-06, |
| "loss": 0.8894, |
| "num_input_tokens_seen": 10592, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 8.230071067810059, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 1.0598, |
| "num_input_tokens_seen": 13856, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.43859649122807015, |
| "grad_norm": 8.494680404663086, |
| "learning_rate": 1.0526315789473684e-05, |
| "loss": 0.688, |
| "num_input_tokens_seen": 17760, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 3.150935649871826, |
| "learning_rate": 1.2719298245614037e-05, |
| "loss": 0.4389, |
| "num_input_tokens_seen": 20704, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6140350877192983, |
| "grad_norm": 8.861291885375977, |
| "learning_rate": 1.4912280701754386e-05, |
| "loss": 0.2838, |
| "num_input_tokens_seen": 23392, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 1.700042963027954, |
| "learning_rate": 1.7105263157894737e-05, |
| "loss": 0.2231, |
| "num_input_tokens_seen": 26176, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 1.2070605754852295, |
| "learning_rate": 1.929824561403509e-05, |
| "loss": 0.2133, |
| "num_input_tokens_seen": 29536, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 1.4389554262161255, |
| "learning_rate": 2.149122807017544e-05, |
| "loss": 0.1543, |
| "num_input_tokens_seen": 32704, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9649122807017544, |
| "grad_norm": 0.6316112875938416, |
| "learning_rate": 2.368421052631579e-05, |
| "loss": 0.1849, |
| "num_input_tokens_seen": 35392, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.220956489443779, |
| "eval_runtime": 0.6177, |
| "eval_samples_per_second": 40.474, |
| "eval_steps_per_second": 11.333, |
| "num_input_tokens_seen": 36112, |
| "step": 57 |
| }, |
| { |
| "epoch": 1.0526315789473684, |
| "grad_norm": 3.5664875507354736, |
| "learning_rate": 2.5877192982456143e-05, |
| "loss": 0.4888, |
| "num_input_tokens_seen": 38192, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1403508771929824, |
| "grad_norm": 1.8096601963043213, |
| "learning_rate": 2.8070175438596492e-05, |
| "loss": 0.2132, |
| "num_input_tokens_seen": 41424, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.2280701754385965, |
| "grad_norm": 0.9523677229881287, |
| "learning_rate": 3.0263157894736844e-05, |
| "loss": 0.1007, |
| "num_input_tokens_seen": 44496, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.3157894736842106, |
| "grad_norm": 0.6207419037818909, |
| "learning_rate": 3.24561403508772e-05, |
| "loss": 0.1902, |
| "num_input_tokens_seen": 47408, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.4035087719298245, |
| "grad_norm": 3.3229734897613525, |
| "learning_rate": 3.4649122807017546e-05, |
| "loss": 0.2065, |
| "num_input_tokens_seen": 50256, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.4912280701754386, |
| "grad_norm": 2.6619460582733154, |
| "learning_rate": 3.6842105263157895e-05, |
| "loss": 0.0858, |
| "num_input_tokens_seen": 52816, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.5789473684210527, |
| "grad_norm": 0.4525831341743469, |
| "learning_rate": 3.9035087719298244e-05, |
| "loss": 0.1913, |
| "num_input_tokens_seen": 55856, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 2.3102519512176514, |
| "learning_rate": 4.12280701754386e-05, |
| "loss": 0.0456, |
| "num_input_tokens_seen": 59920, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.7543859649122808, |
| "grad_norm": 0.5295023918151855, |
| "learning_rate": 4.342105263157895e-05, |
| "loss": 0.0872, |
| "num_input_tokens_seen": 62512, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.8421052631578947, |
| "grad_norm": 5.770318031311035, |
| "learning_rate": 4.56140350877193e-05, |
| "loss": 0.1139, |
| "num_input_tokens_seen": 66288, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.9298245614035088, |
| "grad_norm": 0.052069008350372314, |
| "learning_rate": 4.780701754385965e-05, |
| "loss": 0.2067, |
| "num_input_tokens_seen": 69328, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.2627955973148346, |
| "eval_runtime": 0.6248, |
| "eval_samples_per_second": 40.015, |
| "eval_steps_per_second": 11.204, |
| "num_input_tokens_seen": 71552, |
| "step": 114 |
| }, |
| { |
| "epoch": 2.017543859649123, |
| "grad_norm": 4.355669975280762, |
| "learning_rate": 5e-05, |
| "loss": 0.0985, |
| "num_input_tokens_seen": 71968, |
| "step": 115 |
| }, |
| { |
| "epoch": 2.1052631578947367, |
| "grad_norm": 0.018558036535978317, |
| "learning_rate": 4.999707014206475e-05, |
| "loss": 0.0652, |
| "num_input_tokens_seen": 75040, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.192982456140351, |
| "grad_norm": 2.251081943511963, |
| "learning_rate": 4.9988281254984414e-05, |
| "loss": 0.1731, |
| "num_input_tokens_seen": 78624, |
| "step": 125 |
| }, |
| { |
| "epoch": 2.280701754385965, |
| "grad_norm": 0.14908963441848755, |
| "learning_rate": 4.997363539877422e-05, |
| "loss": 0.1031, |
| "num_input_tokens_seen": 81824, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.3684210526315788, |
| "grad_norm": 0.1386968195438385, |
| "learning_rate": 4.9953136006256415e-05, |
| "loss": 0.0068, |
| "num_input_tokens_seen": 84960, |
| "step": 135 |
| }, |
| { |
| "epoch": 2.456140350877193, |
| "grad_norm": 3.9500904083251953, |
| "learning_rate": 4.9926787882255636e-05, |
| "loss": 0.1244, |
| "num_input_tokens_seen": 89024, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.543859649122807, |
| "grad_norm": 0.17092008888721466, |
| "learning_rate": 4.9894597202472696e-05, |
| "loss": 0.0115, |
| "num_input_tokens_seen": 92480, |
| "step": 145 |
| }, |
| { |
| "epoch": 2.6315789473684212, |
| "grad_norm": 0.01844712346792221, |
| "learning_rate": 4.985657151203706e-05, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 95520, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.719298245614035, |
| "grad_norm": 0.04460814967751503, |
| "learning_rate": 4.9812719723738435e-05, |
| "loss": 0.0973, |
| "num_input_tokens_seen": 98976, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.807017543859649, |
| "grad_norm": 5.250396251678467, |
| "learning_rate": 4.976305211593758e-05, |
| "loss": 0.1626, |
| "num_input_tokens_seen": 101600, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.8947368421052633, |
| "grad_norm": 0.048174094408750534, |
| "learning_rate": 4.970758033015731e-05, |
| "loss": 0.0129, |
| "num_input_tokens_seen": 104544, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.982456140350877, |
| "grad_norm": 8.45261001586914, |
| "learning_rate": 4.9646317368353743e-05, |
| "loss": 0.2322, |
| "num_input_tokens_seen": 107936, |
| "step": 170 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.09871465712785721, |
| "eval_runtime": 0.6314, |
| "eval_samples_per_second": 39.592, |
| "eval_steps_per_second": 11.086, |
| "num_input_tokens_seen": 108088, |
| "step": 171 |
| }, |
| { |
| "epoch": 3.0701754385964914, |
| "grad_norm": 2.33796763420105, |
| "learning_rate": 4.957927758986888e-05, |
| "loss": 0.0094, |
| "num_input_tokens_seen": 110360, |
| "step": 175 |
| }, |
| { |
| "epoch": 3.1578947368421053, |
| "grad_norm": 0.07165002077817917, |
| "learning_rate": 4.9506476708064865e-05, |
| "loss": 0.1153, |
| "num_input_tokens_seen": 113656, |
| "step": 180 |
| }, |
| { |
| "epoch": 3.245614035087719, |
| "grad_norm": 0.3638395369052887, |
| "learning_rate": 4.9427931786641e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 116792, |
| "step": 185 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.03217436373233795, |
| "learning_rate": 4.93436612356342e-05, |
| "loss": 0.0619, |
| "num_input_tokens_seen": 119864, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.4210526315789473, |
| "grad_norm": 0.07817409187555313, |
| "learning_rate": 4.925368480710385e-05, |
| "loss": 0.09, |
| "num_input_tokens_seen": 123512, |
| "step": 195 |
| }, |
| { |
| "epoch": 3.5087719298245617, |
| "grad_norm": 1.4869526624679565, |
| "learning_rate": 4.915802359050222e-05, |
| "loss": 0.051, |
| "num_input_tokens_seen": 127064, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.5964912280701755, |
| "grad_norm": 0.6404334306716919, |
| "learning_rate": 4.905670000773126e-05, |
| "loss": 0.0517, |
| "num_input_tokens_seen": 130008, |
| "step": 205 |
| }, |
| { |
| "epoch": 3.6842105263157894, |
| "grad_norm": 0.04731073975563049, |
| "learning_rate": 4.894973780788722e-05, |
| "loss": 0.0138, |
| "num_input_tokens_seen": 132760, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.7719298245614032, |
| "grad_norm": 0.5577517747879028, |
| "learning_rate": 4.88371620616941e-05, |
| "loss": 0.0877, |
| "num_input_tokens_seen": 136728, |
| "step": 215 |
| }, |
| { |
| "epoch": 3.8596491228070176, |
| "grad_norm": 0.020540254190564156, |
| "learning_rate": 4.871899915562736e-05, |
| "loss": 0.1713, |
| "num_input_tokens_seen": 139800, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.9473684210526314, |
| "grad_norm": 7.120255470275879, |
| "learning_rate": 4.8595276785729236e-05, |
| "loss": 0.1315, |
| "num_input_tokens_seen": 143128, |
| "step": 225 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.1422051042318344, |
| "eval_runtime": 0.6252, |
| "eval_samples_per_second": 39.986, |
| "eval_steps_per_second": 11.196, |
| "num_input_tokens_seen": 144720, |
| "step": 228 |
| }, |
| { |
| "epoch": 4.035087719298246, |
| "grad_norm": 1.2261435985565186, |
| "learning_rate": 4.846602395111711e-05, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 145904, |
| "step": 230 |
| }, |
| { |
| "epoch": 4.12280701754386, |
| "grad_norm": 10.433084487915039, |
| "learning_rate": 4.833127094718643e-05, |
| "loss": 0.0643, |
| "num_input_tokens_seen": 149520, |
| "step": 235 |
| }, |
| { |
| "epoch": 4.2105263157894735, |
| "grad_norm": 0.5427061319351196, |
| "learning_rate": 4.819104935850983e-05, |
| "loss": 0.0022, |
| "num_input_tokens_seen": 152912, |
| "step": 240 |
| }, |
| { |
| "epoch": 4.298245614035087, |
| "grad_norm": 0.013390785083174706, |
| "learning_rate": 4.804539205143405e-05, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 156176, |
| "step": 245 |
| }, |
| { |
| "epoch": 4.385964912280702, |
| "grad_norm": 0.4203593134880066, |
| "learning_rate": 4.789433316637644e-05, |
| "loss": 0.0039, |
| "num_input_tokens_seen": 159376, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.473684210526316, |
| "grad_norm": 5.387624740600586, |
| "learning_rate": 4.7737908109822854e-05, |
| "loss": 0.1729, |
| "num_input_tokens_seen": 162128, |
| "step": 255 |
| }, |
| { |
| "epoch": 4.56140350877193, |
| "grad_norm": 0.1757950335741043, |
| "learning_rate": 4.757615354602874e-05, |
| "loss": 0.0298, |
| "num_input_tokens_seen": 165456, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.649122807017544, |
| "grad_norm": 0.4920128285884857, |
| "learning_rate": 4.7409107388425504e-05, |
| "loss": 0.0041, |
| "num_input_tokens_seen": 168272, |
| "step": 265 |
| }, |
| { |
| "epoch": 4.7368421052631575, |
| "grad_norm": 8.443994522094727, |
| "learning_rate": 4.723680879073396e-05, |
| "loss": 0.0183, |
| "num_input_tokens_seen": 171216, |
| "step": 270 |
| }, |
| { |
| "epoch": 4.824561403508772, |
| "grad_norm": 0.05802755430340767, |
| "learning_rate": 4.70592981377872e-05, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 174192, |
| "step": 275 |
| }, |
| { |
| "epoch": 4.912280701754386, |
| "grad_norm": 0.017686259001493454, |
| "learning_rate": 4.6876617036064844e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 178320, |
| "step": 280 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.3446243405342102, |
| "learning_rate": 4.668880830394093e-05, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 181120, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.08571169525384903, |
| "eval_runtime": 0.6268, |
| "eval_samples_per_second": 39.888, |
| "eval_steps_per_second": 11.169, |
| "num_input_tokens_seen": 181120, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.087719298245614, |
| "grad_norm": 0.13619312644004822, |
| "learning_rate": 4.649591596164778e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 184224, |
| "step": 290 |
| }, |
| { |
| "epoch": 5.175438596491228, |
| "grad_norm": 0.0069135138764977455, |
| "learning_rate": 4.629798522095818e-05, |
| "loss": 0.0024, |
| "num_input_tokens_seen": 187296, |
| "step": 295 |
| }, |
| { |
| "epoch": 5.2631578947368425, |
| "grad_norm": 0.012735686264932156, |
| "learning_rate": 4.6095062474588225e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 190944, |
| "step": 300 |
| }, |
| { |
| "epoch": 5.350877192982456, |
| "grad_norm": 0.003043313743546605, |
| "learning_rate": 4.588719528532342e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 194080, |
| "step": 305 |
| }, |
| { |
| "epoch": 5.43859649122807, |
| "grad_norm": 0.007411586586385965, |
| "learning_rate": 4.5674432374870455e-05, |
| "loss": 0.058, |
| "num_input_tokens_seen": 197312, |
| "step": 310 |
| }, |
| { |
| "epoch": 5.526315789473684, |
| "grad_norm": 0.2823081612586975, |
| "learning_rate": 4.545682361243748e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 201280, |
| "step": 315 |
| }, |
| { |
| "epoch": 5.614035087719298, |
| "grad_norm": 0.0030117405112832785, |
| "learning_rate": 4.5234420003045236e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 204256, |
| "step": 320 |
| }, |
| { |
| "epoch": 5.701754385964913, |
| "grad_norm": 0.09608283638954163, |
| "learning_rate": 4.5007273675572104e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 207488, |
| "step": 325 |
| }, |
| { |
| "epoch": 5.7894736842105265, |
| "grad_norm": 0.001678548171184957, |
| "learning_rate": 4.4775437870535685e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 210240, |
| "step": 330 |
| }, |
| { |
| "epoch": 5.87719298245614, |
| "grad_norm": 0.5041198134422302, |
| "learning_rate": 4.4538966927613836e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 213952, |
| "step": 335 |
| }, |
| { |
| "epoch": 5.964912280701754, |
| "grad_norm": 0.0011690068058669567, |
| "learning_rate": 4.4297916272908024e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 216352, |
| "step": 340 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.052279453724622726, |
| "eval_runtime": 0.6295, |
| "eval_samples_per_second": 39.712, |
| "eval_steps_per_second": 11.119, |
| "num_input_tokens_seen": 217128, |
| "step": 342 |
| }, |
| { |
| "epoch": 6.052631578947368, |
| "grad_norm": 0.010593375191092491, |
| "learning_rate": 4.405234240595214e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 219528, |
| "step": 345 |
| }, |
| { |
| "epoch": 6.140350877192983, |
| "grad_norm": 0.05453763157129288, |
| "learning_rate": 4.3802302886469606e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 222472, |
| "step": 350 |
| }, |
| { |
| "epoch": 6.228070175438597, |
| "grad_norm": 0.008343326859176159, |
| "learning_rate": 4.3547856320882044e-05, |
| "loss": 0.0067, |
| "num_input_tokens_seen": 225448, |
| "step": 355 |
| }, |
| { |
| "epoch": 6.315789473684211, |
| "grad_norm": 3.636948347091675, |
| "learning_rate": 4.328906234857259e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 228584, |
| "step": 360 |
| }, |
| { |
| "epoch": 6.4035087719298245, |
| "grad_norm": 0.005142468959093094, |
| "learning_rate": 4.302598162790712e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 231624, |
| "step": 365 |
| }, |
| { |
| "epoch": 6.491228070175438, |
| "grad_norm": 0.001550496555864811, |
| "learning_rate": 4.27586758220166e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 234856, |
| "step": 370 |
| }, |
| { |
| "epoch": 6.578947368421053, |
| "grad_norm": 0.003545548766851425, |
| "learning_rate": 4.2487207584343955e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 238024, |
| "step": 375 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.009749687276780605, |
| "learning_rate": 4.2211640543958796e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 241576, |
| "step": 380 |
| }, |
| { |
| "epoch": 6.754385964912281, |
| "grad_norm": 0.0016492783324792981, |
| "learning_rate": 4.193203929064353e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 244680, |
| "step": 385 |
| }, |
| { |
| "epoch": 6.842105263157895, |
| "grad_norm": 106.49560546875, |
| "learning_rate": 4.164846935975421e-05, |
| "loss": 0.0489, |
| "num_input_tokens_seen": 247880, |
| "step": 390 |
| }, |
| { |
| "epoch": 6.9298245614035086, |
| "grad_norm": 0.0007353152614086866, |
| "learning_rate": 4.136099721685983e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 251496, |
| "step": 395 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.04707249999046326, |
| "eval_runtime": 0.6289, |
| "eval_samples_per_second": 39.75, |
| "eval_steps_per_second": 11.13, |
| "num_input_tokens_seen": 253536, |
| "step": 399 |
| }, |
| { |
| "epoch": 7.017543859649122, |
| "grad_norm": 0.0012180983321741223, |
| "learning_rate": 4.1069690242163484e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 254016, |
| "step": 400 |
| }, |
| { |
| "epoch": 7.105263157894737, |
| "grad_norm": 0.0014070946490392089, |
| "learning_rate": 4.0774616714709316e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 257024, |
| "step": 405 |
| }, |
| { |
| "epoch": 7.192982456140351, |
| "grad_norm": 0.0011609707726165652, |
| "learning_rate": 4.047584579637857e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 260096, |
| "step": 410 |
| }, |
| { |
| "epoch": 7.280701754385965, |
| "grad_norm": 0.0011609778739511967, |
| "learning_rate": 4.0173447515678916e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 263776, |
| "step": 415 |
| }, |
| { |
| "epoch": 7.368421052631579, |
| "grad_norm": 0.7137589454650879, |
| "learning_rate": 3.986749275133057e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 267008, |
| "step": 420 |
| }, |
| { |
| "epoch": 7.456140350877193, |
| "grad_norm": 0.0011328003602102399, |
| "learning_rate": 3.955805321565304e-05, |
| "loss": 0.0509, |
| "num_input_tokens_seen": 270368, |
| "step": 425 |
| }, |
| { |
| "epoch": 7.543859649122807, |
| "grad_norm": 0.003327790880575776, |
| "learning_rate": 3.9245201437756654e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 273568, |
| "step": 430 |
| }, |
| { |
| "epoch": 7.631578947368421, |
| "grad_norm": 0.0018149224342778325, |
| "learning_rate": 3.892901074654255e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 277152, |
| "step": 435 |
| }, |
| { |
| "epoch": 7.719298245614035, |
| "grad_norm": 0.0024186996743083, |
| "learning_rate": 3.860955525351516e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 280640, |
| "step": 440 |
| }, |
| { |
| "epoch": 7.807017543859649, |
| "grad_norm": 0.0034955486189574003, |
| "learning_rate": 3.82869098354114e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 283392, |
| "step": 445 |
| }, |
| { |
| "epoch": 7.894736842105263, |
| "grad_norm": 0.000977631425485015, |
| "learning_rate": 3.796115011665034e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 286496, |
| "step": 450 |
| }, |
| { |
| "epoch": 7.982456140350877, |
| "grad_norm": 0.00181960454210639, |
| "learning_rate": 3.763235245160775e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 289952, |
| "step": 455 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.008637478575110435, |
| "eval_runtime": 0.6247, |
| "eval_samples_per_second": 40.021, |
| "eval_steps_per_second": 11.206, |
| "num_input_tokens_seen": 290112, |
| "step": 456 |
| }, |
| { |
| "epoch": 8.070175438596491, |
| "grad_norm": 0.0010766127379611135, |
| "learning_rate": 3.7300593906719464e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 292512, |
| "step": 460 |
| }, |
| { |
| "epoch": 8.157894736842104, |
| "grad_norm": 60.37736511230469, |
| "learning_rate": 3.69659522424179e-05, |
| "loss": 0.0188, |
| "num_input_tokens_seen": 295456, |
| "step": 465 |
| }, |
| { |
| "epoch": 8.24561403508772, |
| "grad_norm": 0.0011705532670021057, |
| "learning_rate": 3.662850589490592e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 299424, |
| "step": 470 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.0013269799528643489, |
| "learning_rate": 3.628833395777224e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 302656, |
| "step": 475 |
| }, |
| { |
| "epoch": 8.421052631578947, |
| "grad_norm": 0.0009511318639852107, |
| "learning_rate": 3.59455161634528e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 305952, |
| "step": 480 |
| }, |
| { |
| "epoch": 8.508771929824562, |
| "grad_norm": 0.001095997984521091, |
| "learning_rate": 3.560013286454242e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 309248, |
| "step": 485 |
| }, |
| { |
| "epoch": 8.596491228070175, |
| "grad_norm": 0.005922127515077591, |
| "learning_rate": 3.5252265014961006e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 312224, |
| "step": 490 |
| }, |
| { |
| "epoch": 8.68421052631579, |
| "grad_norm": 0.00104831054341048, |
| "learning_rate": 3.490199415097892e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 315200, |
| "step": 495 |
| }, |
| { |
| "epoch": 8.771929824561404, |
| "grad_norm": 0.0013078617630526423, |
| "learning_rate": 3.45494023721058e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 318240, |
| "step": 500 |
| }, |
| { |
| "epoch": 8.859649122807017, |
| "grad_norm": 0.0006194747402332723, |
| "learning_rate": 3.4194572321847336e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 321408, |
| "step": 505 |
| }, |
| { |
| "epoch": 8.947368421052632, |
| "grad_norm": 0.000688323169015348, |
| "learning_rate": 3.383758716833459e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 324800, |
| "step": 510 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.07975656539201736, |
| "eval_runtime": 0.6288, |
| "eval_samples_per_second": 39.757, |
| "eval_steps_per_second": 11.132, |
| "num_input_tokens_seen": 325872, |
| "step": 513 |
| }, |
| { |
| "epoch": 9.035087719298245, |
| "grad_norm": 0.0006226343102753162, |
| "learning_rate": 3.347853058483037e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 327184, |
| "step": 515 |
| }, |
| { |
| "epoch": 9.12280701754386, |
| "grad_norm": 0.0007310142391361296, |
| "learning_rate": 3.311748673011709e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 329744, |
| "step": 520 |
| }, |
| { |
| "epoch": 9.210526315789474, |
| "grad_norm": 0.010815474204719067, |
| "learning_rate": 3.275454022877097e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 332976, |
| "step": 525 |
| }, |
| { |
| "epoch": 9.298245614035087, |
| "grad_norm": 0.0007622938137501478, |
| "learning_rate": 3.238977615132697e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 335760, |
| "step": 530 |
| }, |
| { |
| "epoch": 9.385964912280702, |
| "grad_norm": 0.0007046781247481704, |
| "learning_rate": 3.202327999433924e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 338128, |
| "step": 535 |
| }, |
| { |
| "epoch": 9.473684210526315, |
| "grad_norm": 0.0005359944188967347, |
| "learning_rate": 3.165513766034167e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 342160, |
| "step": 540 |
| }, |
| { |
| "epoch": 9.56140350877193, |
| "grad_norm": 0.0023622806183993816, |
| "learning_rate": 3.128543543771336e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 344976, |
| "step": 545 |
| }, |
| { |
| "epoch": 9.649122807017545, |
| "grad_norm": 0.0010457730386406183, |
| "learning_rate": 3.091425998045356e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 348208, |
| "step": 550 |
| }, |
| { |
| "epoch": 9.736842105263158, |
| "grad_norm": 0.0035423182416707277, |
| "learning_rate": 3.0541698287870965e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 351472, |
| "step": 555 |
| }, |
| { |
| "epoch": 9.824561403508772, |
| "grad_norm": 0.0007882198551669717, |
| "learning_rate": 3.01678376841921e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 355152, |
| "step": 560 |
| }, |
| { |
| "epoch": 9.912280701754385, |
| "grad_norm": 0.0012028486235067248, |
| "learning_rate": 2.9792765798093465e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 359184, |
| "step": 565 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.0005414988263510168, |
| "learning_rate": 2.94165705421624e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 361920, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.08659715205430984, |
| "eval_runtime": 0.6267, |
| "eval_samples_per_second": 39.894, |
| "eval_steps_per_second": 11.17, |
| "num_input_tokens_seen": 361920, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.087719298245615, |
| "grad_norm": 0.0016271622152999043, |
| "learning_rate": 2.9039340092291373e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 365216, |
| "step": 575 |
| }, |
| { |
| "epoch": 10.175438596491228, |
| "grad_norm": 0.0016625354764983058, |
| "learning_rate": 2.8661162867010543e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 368512, |
| "step": 580 |
| }, |
| { |
| "epoch": 10.263157894736842, |
| "grad_norm": 0.0011906063882634044, |
| "learning_rate": 2.8282127506763456e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 372032, |
| "step": 585 |
| }, |
| { |
| "epoch": 10.350877192982455, |
| "grad_norm": 0.0005975685198791325, |
| "learning_rate": 2.7902322853130757e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 375136, |
| "step": 590 |
| }, |
| { |
| "epoch": 10.43859649122807, |
| "grad_norm": 0.0006688728462904692, |
| "learning_rate": 2.752183792800671e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 378208, |
| "step": 595 |
| }, |
| { |
| "epoch": 10.526315789473685, |
| "grad_norm": 0.0006465315818786621, |
| "learning_rate": 2.7140761912733474e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 380864, |
| "step": 600 |
| }, |
| { |
| "epoch": 10.614035087719298, |
| "grad_norm": 0.0009573330171406269, |
| "learning_rate": 2.6759184127198046e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 384160, |
| "step": 605 |
| }, |
| { |
| "epoch": 10.701754385964913, |
| "grad_norm": 0.0007978378562256694, |
| "learning_rate": 2.6377194008896637e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 387584, |
| "step": 610 |
| }, |
| { |
| "epoch": 10.789473684210526, |
| "grad_norm": 0.0008306215167976916, |
| "learning_rate": 2.5994881091971605e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 391584, |
| "step": 615 |
| }, |
| { |
| "epoch": 10.87719298245614, |
| "grad_norm": 0.0006374895456247032, |
| "learning_rate": 2.5612334986225623e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 394592, |
| "step": 620 |
| }, |
| { |
| "epoch": 10.964912280701755, |
| "grad_norm": 0.0010156974894925952, |
| "learning_rate": 2.5229645356118163e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 397696, |
| "step": 625 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 0.08053161203861237, |
| "eval_runtime": 0.6291, |
| "eval_samples_per_second": 39.74, |
| "eval_steps_per_second": 11.127, |
| "num_input_tokens_seen": 398432, |
| "step": 627 |
| }, |
| { |
| "epoch": 11.052631578947368, |
| "grad_norm": 0.001555000082589686, |
| "learning_rate": 2.4846901899749185e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 400736, |
| "step": 630 |
| }, |
| { |
| "epoch": 11.140350877192983, |
| "grad_norm": 0.0004488785343710333, |
| "learning_rate": 2.4464194327834926e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 403680, |
| "step": 635 |
| }, |
| { |
| "epoch": 11.228070175438596, |
| "grad_norm": 0.0007279742858372629, |
| "learning_rate": 2.4081612342680694e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 406784, |
| "step": 640 |
| }, |
| { |
| "epoch": 11.31578947368421, |
| "grad_norm": 0.0006254904437810183, |
| "learning_rate": 2.369924561715569e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 410208, |
| "step": 645 |
| }, |
| { |
| "epoch": 11.403508771929825, |
| "grad_norm": 0.0006179868942126632, |
| "learning_rate": 2.3317183773674718e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 414016, |
| "step": 650 |
| }, |
| { |
| "epoch": 11.491228070175438, |
| "grad_norm": 0.0007071035215631127, |
| "learning_rate": 2.2935516363191693e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 416992, |
| "step": 655 |
| }, |
| { |
| "epoch": 11.578947368421053, |
| "grad_norm": 0.0007423391216434538, |
| "learning_rate": 2.2554332844209904e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 420224, |
| "step": 660 |
| }, |
| { |
| "epoch": 11.666666666666666, |
| "grad_norm": 0.001821287558414042, |
| "learning_rate": 2.2173722561813987e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 423456, |
| "step": 665 |
| }, |
| { |
| "epoch": 11.75438596491228, |
| "grad_norm": 0.0005227949004620314, |
| "learning_rate": 2.179377472672842e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 426560, |
| "step": 670 |
| }, |
| { |
| "epoch": 11.842105263157894, |
| "grad_norm": 0.0009789641480892897, |
| "learning_rate": 2.1414578394407597e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 429920, |
| "step": 675 |
| }, |
| { |
| "epoch": 11.929824561403509, |
| "grad_norm": 0.0005592002999037504, |
| "learning_rate": 2.1036222444162147e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 433088, |
| "step": 680 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 0.08902089297771454, |
| "eval_runtime": 0.6286, |
| "eval_samples_per_second": 39.771, |
| "eval_steps_per_second": 11.136, |
| "num_input_tokens_seen": 435536, |
| "step": 684 |
| }, |
| { |
| "epoch": 12.017543859649123, |
| "grad_norm": 0.0009320858516730368, |
| "learning_rate": 2.0658795558326743e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 436304, |
| "step": 685 |
| }, |
| { |
| "epoch": 12.105263157894736, |
| "grad_norm": 0.0010096331825479865, |
| "learning_rate": 2.0282386201473894e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 439472, |
| "step": 690 |
| }, |
| { |
| "epoch": 12.192982456140351, |
| "grad_norm": 0.001230765599757433, |
| "learning_rate": 1.99070825996789e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 442160, |
| "step": 695 |
| }, |
| { |
| "epoch": 12.280701754385966, |
| "grad_norm": 0.0005884809070266783, |
| "learning_rate": 1.9532972719840607e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 445488, |
| "step": 700 |
| }, |
| { |
| "epoch": 12.368421052631579, |
| "grad_norm": 0.0005540026468224823, |
| "learning_rate": 1.9160144249063035e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 448784, |
| "step": 705 |
| }, |
| { |
| "epoch": 12.456140350877194, |
| "grad_norm": 0.0005484636640176177, |
| "learning_rate": 1.8788684574102467e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 452656, |
| "step": 710 |
| }, |
| { |
| "epoch": 12.543859649122806, |
| "grad_norm": 0.0005028496962040663, |
| "learning_rate": 1.8418680760885027e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 455888, |
| "step": 715 |
| }, |
| { |
| "epoch": 12.631578947368421, |
| "grad_norm": 0.0010016572196036577, |
| "learning_rate": 1.805021953409934e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 458896, |
| "step": 720 |
| }, |
| { |
| "epoch": 12.719298245614034, |
| "grad_norm": 0.00038934106123633683, |
| "learning_rate": 1.7683387256869353e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 461648, |
| "step": 725 |
| }, |
| { |
| "epoch": 12.807017543859649, |
| "grad_norm": 0.0028921838384121656, |
| "learning_rate": 1.7318269910511736e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 464912, |
| "step": 730 |
| }, |
| { |
| "epoch": 12.894736842105264, |
| "grad_norm": 0.0007451688870787621, |
| "learning_rate": 1.6954953074382863e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 467888, |
| "step": 735 |
| }, |
| { |
| "epoch": 12.982456140350877, |
| "grad_norm": 0.0005756712635047734, |
| "learning_rate": 1.659352190581993e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 471440, |
| "step": 740 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_loss": 0.08622876554727554, |
| "eval_runtime": 0.6297, |
| "eval_samples_per_second": 39.699, |
| "eval_steps_per_second": 11.116, |
| "num_input_tokens_seen": 471520, |
| "step": 741 |
| }, |
| { |
| "epoch": 13.070175438596491, |
| "grad_norm": 0.0005351323052309453, |
| "learning_rate": 1.6234061120181142e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 473600, |
| "step": 745 |
| }, |
| { |
| "epoch": 13.157894736842104, |
| "grad_norm": 0.001625753939151764, |
| "learning_rate": 1.5876654970989308e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 476576, |
| "step": 750 |
| }, |
| { |
| "epoch": 13.24561403508772, |
| "grad_norm": 0.0004524633113760501, |
| "learning_rate": 1.552138723018382e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 479776, |
| "step": 755 |
| }, |
| { |
| "epoch": 13.333333333333334, |
| "grad_norm": 0.0004668083565775305, |
| "learning_rate": 1.5168341168485423e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 482784, |
| "step": 760 |
| }, |
| { |
| "epoch": 13.421052631578947, |
| "grad_norm": 0.0012625912204384804, |
| "learning_rate": 1.4817599535878565e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 485696, |
| "step": 765 |
| }, |
| { |
| "epoch": 13.508771929824562, |
| "grad_norm": 0.0005987969925627112, |
| "learning_rate": 1.4469244542215682e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 488832, |
| "step": 770 |
| }, |
| { |
| "epoch": 13.596491228070175, |
| "grad_norm": 0.0004978460492566228, |
| "learning_rate": 1.4123357837948175e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 492896, |
| "step": 775 |
| }, |
| { |
| "epoch": 13.68421052631579, |
| "grad_norm": 0.0006448051426559687, |
| "learning_rate": 1.3780020494988446e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 496224, |
| "step": 780 |
| }, |
| { |
| "epoch": 13.771929824561404, |
| "grad_norm": 0.0008532825158908963, |
| "learning_rate": 1.3439312987707615e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 499040, |
| "step": 785 |
| }, |
| { |
| "epoch": 13.859649122807017, |
| "grad_norm": 0.0010070840362459421, |
| "learning_rate": 1.3101315174073162e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 502560, |
| "step": 790 |
| }, |
| { |
| "epoch": 13.947368421052632, |
| "grad_norm": 0.0007303861784748733, |
| "learning_rate": 1.2766106276931223e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 505152, |
| "step": 795 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_loss": 0.08555411547422409, |
| "eval_runtime": 0.6282, |
| "eval_samples_per_second": 39.799, |
| "eval_steps_per_second": 11.144, |
| "num_input_tokens_seen": 507256, |
| "step": 798 |
| }, |
| { |
| "epoch": 14.035087719298245, |
| "grad_norm": 0.0007077206973917782, |
| "learning_rate": 1.243376486543755e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 508888, |
| "step": 800 |
| }, |
| { |
| "epoch": 14.12280701754386, |
| "grad_norm": 0.001242919242940843, |
| "learning_rate": 1.2104368836641908e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 511576, |
| "step": 805 |
| }, |
| { |
| "epoch": 14.210526315789474, |
| "grad_norm": 0.0004700981080532074, |
| "learning_rate": 1.1777995397229771e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 514968, |
| "step": 810 |
| }, |
| { |
| "epoch": 14.298245614035087, |
| "grad_norm": 0.0005096773384138942, |
| "learning_rate": 1.1454721045426073e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 517912, |
| "step": 815 |
| }, |
| { |
| "epoch": 14.385964912280702, |
| "grad_norm": 0.0003988526004832238, |
| "learning_rate": 1.113462155306478e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 521144, |
| "step": 820 |
| }, |
| { |
| "epoch": 14.473684210526315, |
| "grad_norm": 0.0003391786594875157, |
| "learning_rate": 1.0817771947828934e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 524152, |
| "step": 825 |
| }, |
| { |
| "epoch": 14.56140350877193, |
| "grad_norm": 0.0006226114928722382, |
| "learning_rate": 1.0504246495664932e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 526808, |
| "step": 830 |
| }, |
| { |
| "epoch": 14.649122807017545, |
| "grad_norm": 0.0014990364434197545, |
| "learning_rate": 1.0194118683375503e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 530072, |
| "step": 835 |
| }, |
| { |
| "epoch": 14.736842105263158, |
| "grad_norm": 0.0018731486052274704, |
| "learning_rate": 9.887461201395176e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 534104, |
| "step": 840 |
| }, |
| { |
| "epoch": 14.824561403508772, |
| "grad_norm": 0.00041385687654837966, |
| "learning_rate": 9.584345926752524e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 537752, |
| "step": 845 |
| }, |
| { |
| "epoch": 14.912280701754385, |
| "grad_norm": 0.0018344820709899068, |
| "learning_rate": 9.284843906222948e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 540696, |
| "step": 850 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 0.000453752203611657, |
| "learning_rate": 8.98902533967618e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 543064, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.0890091136097908, |
| "eval_runtime": 0.6274, |
| "eval_samples_per_second": 39.85, |
| "eval_steps_per_second": 11.158, |
| "num_input_tokens_seen": 543064, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.087719298245615, |
| "grad_norm": 0.0005497167585417628, |
| "learning_rate": 8.696959563622174e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 546104, |
| "step": 860 |
| }, |
| { |
| "epoch": 15.175438596491228, |
| "grad_norm": 0.0009668822167441249, |
| "learning_rate": 8.40871503495947e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 549112, |
| "step": 865 |
| }, |
| { |
| "epoch": 15.263157894736842, |
| "grad_norm": 0.0004576245555654168, |
| "learning_rate": 8.124359314929622e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 552696, |
| "step": 870 |
| }, |
| { |
| "epoch": 15.350877192982455, |
| "grad_norm": 0.0009820536943152547, |
| "learning_rate": 7.843959053281663e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 556152, |
| "step": 875 |
| }, |
| { |
| "epoch": 15.43859649122807, |
| "grad_norm": 0.001712694764137268, |
| "learning_rate": 7.5675799726501155e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 558936, |
| "step": 880 |
| }, |
| { |
| "epoch": 15.526315789473685, |
| "grad_norm": 0.00041718766442500055, |
| "learning_rate": 7.295286853150391e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 562584, |
| "step": 885 |
| }, |
| { |
| "epoch": 15.614035087719298, |
| "grad_norm": 0.0007184024434536695, |
| "learning_rate": 7.027143517195023e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 566456, |
| "step": 890 |
| }, |
| { |
| "epoch": 15.701754385964913, |
| "grad_norm": 0.0003503764164634049, |
| "learning_rate": 6.763212814534484e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 569176, |
| "step": 895 |
| }, |
| { |
| "epoch": 15.789473684210526, |
| "grad_norm": 0.0005723480717279017, |
| "learning_rate": 6.503556607525838e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 572696, |
| "step": 900 |
| }, |
| { |
| "epoch": 15.87719298245614, |
| "grad_norm": 0.0004324032925069332, |
| "learning_rate": 6.248235756632984e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 575832, |
| "step": 905 |
| }, |
| { |
| "epoch": 15.964912280701755, |
| "grad_norm": 0.0009354011272080243, |
| "learning_rate": 5.997310106161589e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 578808, |
| "step": 910 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_loss": 0.09002669155597687, |
| "eval_runtime": 0.6277, |
| "eval_samples_per_second": 39.83, |
| "eval_steps_per_second": 11.152, |
| "num_input_tokens_seen": 579704, |
| "step": 912 |
| }, |
| { |
| "epoch": 16.05263157894737, |
| "grad_norm": 0.0006182304350659251, |
| "learning_rate": 5.7508384702323226e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 581720, |
| "step": 915 |
| }, |
| { |
| "epoch": 16.140350877192983, |
| "grad_norm": 0.0012547291116788983, |
| "learning_rate": 5.508878618995439e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 585624, |
| "step": 920 |
| }, |
| { |
| "epoch": 16.228070175438596, |
| "grad_norm": 0.0003780796832870692, |
| "learning_rate": 5.271487265090163e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 588408, |
| "step": 925 |
| }, |
| { |
| "epoch": 16.31578947368421, |
| "grad_norm": 0.0028658711817115545, |
| "learning_rate": 5.038720050351842e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 591512, |
| "step": 930 |
| }, |
| { |
| "epoch": 16.403508771929825, |
| "grad_norm": 0.0004841610207222402, |
| "learning_rate": 4.810631532770182e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 595288, |
| "step": 935 |
| }, |
| { |
| "epoch": 16.49122807017544, |
| "grad_norm": 0.0013498651096597314, |
| "learning_rate": 4.587275173701428e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 598360, |
| "step": 940 |
| }, |
| { |
| "epoch": 16.57894736842105, |
| "grad_norm": 0.0014918326633051038, |
| "learning_rate": 4.368703325337667e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 601944, |
| "step": 945 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 0.0005209071096032858, |
| "learning_rate": 4.154967218436037e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 604824, |
| "step": 950 |
| }, |
| { |
| "epoch": 16.75438596491228, |
| "grad_norm": 0.00047317391727119684, |
| "learning_rate": 3.94611695031086e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 607608, |
| "step": 955 |
| }, |
| { |
| "epoch": 16.842105263157894, |
| "grad_norm": 0.00042429487803019583, |
| "learning_rate": 3.74220147309135e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 611128, |
| "step": 960 |
| }, |
| { |
| "epoch": 16.92982456140351, |
| "grad_norm": 0.0008401384693570435, |
| "learning_rate": 3.543268582247844e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 614040, |
| "step": 965 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_loss": 0.09129036962985992, |
| "eval_runtime": 0.6274, |
| "eval_samples_per_second": 39.846, |
| "eval_steps_per_second": 11.157, |
| "num_input_tokens_seen": 615960, |
| "step": 969 |
| }, |
| { |
| "epoch": 17.017543859649123, |
| "grad_norm": 0.0005760001367889345, |
| "learning_rate": 3.3493649053890326e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 616600, |
| "step": 970 |
| }, |
| { |
| "epoch": 17.105263157894736, |
| "grad_norm": 0.00036672543501481414, |
| "learning_rate": 3.1605358913330385e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 619736, |
| "step": 975 |
| }, |
| { |
| "epoch": 17.19298245614035, |
| "grad_norm": 0.0007026438834145665, |
| "learning_rate": 2.9768257994546662e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 622680, |
| "step": 980 |
| }, |
| { |
| "epoch": 17.280701754385966, |
| "grad_norm": 0.001340859686024487, |
| "learning_rate": 2.7982776893115627e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 626168, |
| "step": 985 |
| }, |
| { |
| "epoch": 17.36842105263158, |
| "grad_norm": 0.0006338339881040156, |
| "learning_rate": 2.624933410551508e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 628952, |
| "step": 990 |
| }, |
| { |
| "epoch": 17.45614035087719, |
| "grad_norm": 0.0004151356697548181, |
| "learning_rate": 2.456833593103361e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 632184, |
| "step": 995 |
| }, |
| { |
| "epoch": 17.54385964912281, |
| "grad_norm": 0.0010616648942232132, |
| "learning_rate": 2.2940176376538445e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 635448, |
| "step": 1000 |
| }, |
| { |
| "epoch": 17.63157894736842, |
| "grad_norm": 0.00037299131508916616, |
| "learning_rate": 2.136523706412477e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 638616, |
| "step": 1005 |
| }, |
| { |
| "epoch": 17.719298245614034, |
| "grad_norm": 0.0003296730574220419, |
| "learning_rate": 1.984388714166799e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 641688, |
| "step": 1010 |
| }, |
| { |
| "epoch": 17.80701754385965, |
| "grad_norm": 0.0012227256083860993, |
| "learning_rate": 1.837648319629956e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 644472, |
| "step": 1015 |
| }, |
| { |
| "epoch": 17.894736842105264, |
| "grad_norm": 0.0005031724576838315, |
| "learning_rate": 1.6963369170826943e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 648312, |
| "step": 1020 |
| }, |
| { |
| "epoch": 17.982456140350877, |
| "grad_norm": 0.0036398344673216343, |
| "learning_rate": 1.5604876283117326e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 652184, |
| "step": 1025 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_loss": 0.09119849652051926, |
| "eval_runtime": 0.6291, |
| "eval_samples_per_second": 39.737, |
| "eval_steps_per_second": 11.126, |
| "num_input_tokens_seen": 652368, |
| "step": 1026 |
| }, |
| { |
| "epoch": 18.07017543859649, |
| "grad_norm": 0.0005520170670934021, |
| "learning_rate": 1.4301322948464147e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 655312, |
| "step": 1030 |
| }, |
| { |
| "epoch": 18.157894736842106, |
| "grad_norm": 0.0005950906779617071, |
| "learning_rate": 1.3053014704953987e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 658896, |
| "step": 1035 |
| }, |
| { |
| "epoch": 18.24561403508772, |
| "grad_norm": 0.0006160799530334771, |
| "learning_rate": 1.1860244141851773e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 662288, |
| "step": 1040 |
| }, |
| { |
| "epoch": 18.333333333333332, |
| "grad_norm": 0.00044143779086880386, |
| "learning_rate": 1.0723290831021471e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 665104, |
| "step": 1045 |
| }, |
| { |
| "epoch": 18.42105263157895, |
| "grad_norm": 0.000681056291796267, |
| "learning_rate": 9.642421261397472e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 667536, |
| "step": 1050 |
| }, |
| { |
| "epoch": 18.50877192982456, |
| "grad_norm": 0.00091008614981547, |
| "learning_rate": 8.617888776522642e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 670448, |
| "step": 1055 |
| }, |
| { |
| "epoch": 18.596491228070175, |
| "grad_norm": 0.0018491569207981229, |
| "learning_rate": 7.649933515167407e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 674192, |
| "step": 1060 |
| }, |
| { |
| "epoch": 18.68421052631579, |
| "grad_norm": 0.0009845505701377988, |
| "learning_rate": 6.738782355044049e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 676784, |
| "step": 1065 |
| }, |
| { |
| "epoch": 18.771929824561404, |
| "grad_norm": 0.0009317832300439477, |
| "learning_rate": 5.88464885962911e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 679280, |
| "step": 1070 |
| }, |
| { |
| "epoch": 18.859649122807017, |
| "grad_norm": 0.000517063366714865, |
| "learning_rate": 5.087733228106517e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 682160, |
| "step": 1075 |
| }, |
| { |
| "epoch": 18.94736842105263, |
| "grad_norm": 0.0004094515461474657, |
| "learning_rate": 4.3482222484432513e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 685968, |
| "step": 1080 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_loss": 0.0851987674832344, |
| "eval_runtime": 0.6308, |
| "eval_samples_per_second": 39.633, |
| "eval_steps_per_second": 11.097, |
| "num_input_tokens_seen": 687976, |
| "step": 1083 |
| }, |
| { |
| "epoch": 19.035087719298247, |
| "grad_norm": 0.00043043080950155854, |
| "learning_rate": 3.666289253608235e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 689128, |
| "step": 1085 |
| }, |
| { |
| "epoch": 19.12280701754386, |
| "grad_norm": 0.0029145274311304092, |
| "learning_rate": 3.0420940809451624e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 692136, |
| "step": 1090 |
| }, |
| { |
| "epoch": 19.210526315789473, |
| "grad_norm": 0.000554764992557466, |
| "learning_rate": 2.47578303470844e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 695112, |
| "step": 1095 |
| }, |
| { |
| "epoch": 19.29824561403509, |
| "grad_norm": 0.0004157763614784926, |
| "learning_rate": 1.96748885177106e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 698600, |
| "step": 1100 |
| }, |
| { |
| "epoch": 19.385964912280702, |
| "grad_norm": 0.0044406414963305, |
| "learning_rate": 1.517330670512629e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 701640, |
| "step": 1105 |
| }, |
| { |
| "epoch": 19.473684210526315, |
| "grad_norm": 0.00044373140553943813, |
| "learning_rate": 1.125414002894759e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 704488, |
| "step": 1110 |
| }, |
| { |
| "epoch": 19.56140350877193, |
| "grad_norm": 0.0005462678382173181, |
| "learning_rate": 7.918307097301014e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 707848, |
| "step": 1115 |
| }, |
| { |
| "epoch": 19.649122807017545, |
| "grad_norm": 0.00037007094942964613, |
| "learning_rate": 5.166589791513465e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 711112, |
| "step": 1120 |
| }, |
| { |
| "epoch": 19.736842105263158, |
| "grad_norm": 0.0005305653903633356, |
| "learning_rate": 2.999633082847453e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 714344, |
| "step": 1125 |
| }, |
| { |
| "epoch": 19.82456140350877, |
| "grad_norm": 0.002031900454312563, |
| "learning_rate": 1.4179448813278484e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 718280, |
| "step": 1130 |
| }, |
| { |
| "epoch": 19.912280701754387, |
| "grad_norm": 0.0017279505264014006, |
| "learning_rate": 4.218959166932268e-09, |
| "loss": 0.0, |
| "num_input_tokens_seen": 720968, |
| "step": 1135 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 0.0009181174682453275, |
| "learning_rate": 1.1719651499819683e-10, |
| "loss": 0.0, |
| "num_input_tokens_seen": 723584, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.09011270850896835, |
| "eval_runtime": 0.6285, |
| "eval_samples_per_second": 39.775, |
| "eval_steps_per_second": 11.137, |
| "num_input_tokens_seen": 723584, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "num_input_tokens_seen": 723584, |
| "step": 1140, |
| "total_flos": 3.2646394039762944e+16, |
| "train_loss": 0.04626832362874554, |
| "train_runtime": 330.4495, |
| "train_samples_per_second": 13.618, |
| "train_steps_per_second": 3.45 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1140, |
| "num_input_tokens_seen": 723584, |
| "num_train_epochs": 20, |
| "save_steps": 57, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.2646394039762944e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
