| { |
| "best_global_step": 513, |
| "best_metric": 0.13670699298381805, |
| "best_model_checkpoint": "saves_multiple/ia3/llama-3-8b-instruct/train_cb_789_1760637871/checkpoint-513", |
| "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": 3.3630213737487793, |
| "learning_rate": 1.7543859649122807e-06, |
| "loss": 1.1283, |
| "num_input_tokens_seen": 3136, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 3.312760591506958, |
| "learning_rate": 3.9473684210526315e-06, |
| "loss": 0.9742, |
| "num_input_tokens_seen": 6112, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 2.769252061843872, |
| "learning_rate": 6.140350877192982e-06, |
| "loss": 0.8862, |
| "num_input_tokens_seen": 10112, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 3.3045618534088135, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 1.3094, |
| "num_input_tokens_seen": 13280, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.43859649122807015, |
| "grad_norm": 3.665391206741333, |
| "learning_rate": 1.0526315789473684e-05, |
| "loss": 1.1607, |
| "num_input_tokens_seen": 16288, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 2.7952098846435547, |
| "learning_rate": 1.2719298245614037e-05, |
| "loss": 1.0942, |
| "num_input_tokens_seen": 19104, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6140350877192983, |
| "grad_norm": 2.2919137477874756, |
| "learning_rate": 1.4912280701754386e-05, |
| "loss": 1.1866, |
| "num_input_tokens_seen": 22144, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 3.1309096813201904, |
| "learning_rate": 1.7105263157894737e-05, |
| "loss": 1.002, |
| "num_input_tokens_seen": 25792, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 3.2065813541412354, |
| "learning_rate": 1.929824561403509e-05, |
| "loss": 1.1968, |
| "num_input_tokens_seen": 28576, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 4.102292537689209, |
| "learning_rate": 2.149122807017544e-05, |
| "loss": 1.3349, |
| "num_input_tokens_seen": 31424, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9649122807017544, |
| "grad_norm": 3.059723138809204, |
| "learning_rate": 2.368421052631579e-05, |
| "loss": 0.9866, |
| "num_input_tokens_seen": 34720, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.1527795791625977, |
| "eval_runtime": 0.6542, |
| "eval_samples_per_second": 38.216, |
| "eval_steps_per_second": 10.701, |
| "num_input_tokens_seen": 35448, |
| "step": 57 |
| }, |
| { |
| "epoch": 1.0526315789473684, |
| "grad_norm": 2.330932855606079, |
| "learning_rate": 2.5877192982456143e-05, |
| "loss": 0.8088, |
| "num_input_tokens_seen": 37688, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1403508771929824, |
| "grad_norm": 2.80367374420166, |
| "learning_rate": 2.8070175438596492e-05, |
| "loss": 1.1055, |
| "num_input_tokens_seen": 40792, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.2280701754385965, |
| "grad_norm": 2.9329938888549805, |
| "learning_rate": 3.0263157894736844e-05, |
| "loss": 1.2038, |
| "num_input_tokens_seen": 43832, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.3157894736842106, |
| "grad_norm": 3.0999975204467773, |
| "learning_rate": 3.24561403508772e-05, |
| "loss": 1.0938, |
| "num_input_tokens_seen": 46648, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.4035087719298245, |
| "grad_norm": 3.6106417179107666, |
| "learning_rate": 3.4649122807017546e-05, |
| "loss": 1.0896, |
| "num_input_tokens_seen": 49848, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.4912280701754386, |
| "grad_norm": 3.2087888717651367, |
| "learning_rate": 3.6842105263157895e-05, |
| "loss": 1.2346, |
| "num_input_tokens_seen": 52504, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.5789473684210527, |
| "grad_norm": 3.3049001693725586, |
| "learning_rate": 3.9035087719298244e-05, |
| "loss": 0.9749, |
| "num_input_tokens_seen": 55448, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 3.0932552814483643, |
| "learning_rate": 4.12280701754386e-05, |
| "loss": 1.0834, |
| "num_input_tokens_seen": 58776, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.7543859649122808, |
| "grad_norm": 2.7704927921295166, |
| "learning_rate": 4.342105263157895e-05, |
| "loss": 1.027, |
| "num_input_tokens_seen": 61624, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.8421052631578947, |
| "grad_norm": 2.226386070251465, |
| "learning_rate": 4.56140350877193e-05, |
| "loss": 0.964, |
| "num_input_tokens_seen": 65336, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.9298245614035088, |
| "grad_norm": 2.8043212890625, |
| "learning_rate": 4.780701754385965e-05, |
| "loss": 0.836, |
| "num_input_tokens_seen": 68280, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.5759475827217102, |
| "eval_runtime": 0.6607, |
| "eval_samples_per_second": 37.839, |
| "eval_steps_per_second": 10.595, |
| "num_input_tokens_seen": 70496, |
| "step": 114 |
| }, |
| { |
| "epoch": 2.017543859649123, |
| "grad_norm": 2.1016459465026855, |
| "learning_rate": 5e-05, |
| "loss": 0.5896, |
| "num_input_tokens_seen": 71008, |
| "step": 115 |
| }, |
| { |
| "epoch": 2.1052631578947367, |
| "grad_norm": 3.402238368988037, |
| "learning_rate": 4.999707014206475e-05, |
| "loss": 0.4009, |
| "num_input_tokens_seen": 74176, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.192982456140351, |
| "grad_norm": 1.37735915184021, |
| "learning_rate": 4.9988281254984414e-05, |
| "loss": 0.3185, |
| "num_input_tokens_seen": 77472, |
| "step": 125 |
| }, |
| { |
| "epoch": 2.280701754385965, |
| "grad_norm": 1.1492122411727905, |
| "learning_rate": 4.997363539877422e-05, |
| "loss": 0.2645, |
| "num_input_tokens_seen": 80192, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.3684210526315788, |
| "grad_norm": 1.8137037754058838, |
| "learning_rate": 4.9953136006256415e-05, |
| "loss": 0.1861, |
| "num_input_tokens_seen": 83424, |
| "step": 135 |
| }, |
| { |
| "epoch": 2.456140350877193, |
| "grad_norm": 1.61770761013031, |
| "learning_rate": 4.9926787882255636e-05, |
| "loss": 0.1534, |
| "num_input_tokens_seen": 86080, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.543859649122807, |
| "grad_norm": 3.0257303714752197, |
| "learning_rate": 4.9894597202472696e-05, |
| "loss": 0.2856, |
| "num_input_tokens_seen": 89600, |
| "step": 145 |
| }, |
| { |
| "epoch": 2.6315789473684212, |
| "grad_norm": 1.607239007949829, |
| "learning_rate": 4.985657151203706e-05, |
| "loss": 0.3216, |
| "num_input_tokens_seen": 92704, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.719298245614035, |
| "grad_norm": 0.7693922519683838, |
| "learning_rate": 4.9812719723738435e-05, |
| "loss": 0.1194, |
| "num_input_tokens_seen": 96160, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.807017543859649, |
| "grad_norm": 1.6435840129852295, |
| "learning_rate": 4.976305211593758e-05, |
| "loss": 0.1667, |
| "num_input_tokens_seen": 99520, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.8947368421052633, |
| "grad_norm": 1.592359185218811, |
| "learning_rate": 4.970758033015731e-05, |
| "loss": 0.1479, |
| "num_input_tokens_seen": 102528, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.982456140350877, |
| "grad_norm": 0.4187203049659729, |
| "learning_rate": 4.9646317368353743e-05, |
| "loss": 0.1908, |
| "num_input_tokens_seen": 106304, |
| "step": 170 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.169634148478508, |
| "eval_runtime": 0.66, |
| "eval_samples_per_second": 37.879, |
| "eval_steps_per_second": 10.606, |
| "num_input_tokens_seen": 106416, |
| "step": 171 |
| }, |
| { |
| "epoch": 3.0701754385964914, |
| "grad_norm": 1.0153694152832031, |
| "learning_rate": 4.957927758986888e-05, |
| "loss": 0.0832, |
| "num_input_tokens_seen": 108880, |
| "step": 175 |
| }, |
| { |
| "epoch": 3.1578947368421053, |
| "grad_norm": 1.7011926174163818, |
| "learning_rate": 4.9506476708064865e-05, |
| "loss": 0.1749, |
| "num_input_tokens_seen": 111984, |
| "step": 180 |
| }, |
| { |
| "epoch": 3.245614035087719, |
| "grad_norm": 0.29407209157943726, |
| "learning_rate": 4.9427931786641e-05, |
| "loss": 0.0705, |
| "num_input_tokens_seen": 115248, |
| "step": 185 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 2.0715649127960205, |
| "learning_rate": 4.93436612356342e-05, |
| "loss": 0.0819, |
| "num_input_tokens_seen": 118384, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.4210526315789473, |
| "grad_norm": 0.6385584473609924, |
| "learning_rate": 4.925368480710385e-05, |
| "loss": 0.1212, |
| "num_input_tokens_seen": 122224, |
| "step": 195 |
| }, |
| { |
| "epoch": 3.5087719298245617, |
| "grad_norm": 0.7050696611404419, |
| "learning_rate": 4.915802359050222e-05, |
| "loss": 0.1123, |
| "num_input_tokens_seen": 125584, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.5964912280701755, |
| "grad_norm": 0.36255767941474915, |
| "learning_rate": 4.905670000773126e-05, |
| "loss": 0.0967, |
| "num_input_tokens_seen": 128816, |
| "step": 205 |
| }, |
| { |
| "epoch": 3.6842105263157894, |
| "grad_norm": 2.45294189453125, |
| "learning_rate": 4.894973780788722e-05, |
| "loss": 0.4245, |
| "num_input_tokens_seen": 131952, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.7719298245614032, |
| "grad_norm": 0.09704189002513885, |
| "learning_rate": 4.88371620616941e-05, |
| "loss": 0.1841, |
| "num_input_tokens_seen": 135248, |
| "step": 215 |
| }, |
| { |
| "epoch": 3.8596491228070176, |
| "grad_norm": 0.05094490945339203, |
| "learning_rate": 4.871899915562736e-05, |
| "loss": 0.0577, |
| "num_input_tokens_seen": 138384, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.9473684210526314, |
| "grad_norm": 3.1351993083953857, |
| "learning_rate": 4.8595276785729236e-05, |
| "loss": 0.3003, |
| "num_input_tokens_seen": 141264, |
| "step": 225 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.14660418033599854, |
| "eval_runtime": 0.6618, |
| "eval_samples_per_second": 37.775, |
| "eval_steps_per_second": 10.577, |
| "num_input_tokens_seen": 142480, |
| "step": 228 |
| }, |
| { |
| "epoch": 4.035087719298246, |
| "grad_norm": 1.2932846546173096, |
| "learning_rate": 4.846602395111711e-05, |
| "loss": 0.2522, |
| "num_input_tokens_seen": 143664, |
| "step": 230 |
| }, |
| { |
| "epoch": 4.12280701754386, |
| "grad_norm": 0.20881390571594238, |
| "learning_rate": 4.833127094718643e-05, |
| "loss": 0.0566, |
| "num_input_tokens_seen": 146800, |
| "step": 235 |
| }, |
| { |
| "epoch": 4.2105263157894735, |
| "grad_norm": 0.47832801938056946, |
| "learning_rate": 4.819104935850983e-05, |
| "loss": 0.0631, |
| "num_input_tokens_seen": 149456, |
| "step": 240 |
| }, |
| { |
| "epoch": 4.298245614035087, |
| "grad_norm": 0.16718034446239471, |
| "learning_rate": 4.804539205143405e-05, |
| "loss": 0.2271, |
| "num_input_tokens_seen": 152880, |
| "step": 245 |
| }, |
| { |
| "epoch": 4.385964912280702, |
| "grad_norm": 0.3414199948310852, |
| "learning_rate": 4.789433316637644e-05, |
| "loss": 0.0946, |
| "num_input_tokens_seen": 156720, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.473684210526316, |
| "grad_norm": 0.3329034149646759, |
| "learning_rate": 4.7737908109822854e-05, |
| "loss": 0.1502, |
| "num_input_tokens_seen": 159600, |
| "step": 255 |
| }, |
| { |
| "epoch": 4.56140350877193, |
| "grad_norm": 1.7564311027526855, |
| "learning_rate": 4.757615354602874e-05, |
| "loss": 0.1136, |
| "num_input_tokens_seen": 162224, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.649122807017544, |
| "grad_norm": 0.6217461824417114, |
| "learning_rate": 4.7409107388425504e-05, |
| "loss": 0.0807, |
| "num_input_tokens_seen": 165328, |
| "step": 265 |
| }, |
| { |
| "epoch": 4.7368421052631575, |
| "grad_norm": 2.205641031265259, |
| "learning_rate": 4.723680879073396e-05, |
| "loss": 0.1521, |
| "num_input_tokens_seen": 168592, |
| "step": 270 |
| }, |
| { |
| "epoch": 4.824561403508772, |
| "grad_norm": 0.15078698098659515, |
| "learning_rate": 4.70592981377872e-05, |
| "loss": 0.1895, |
| "num_input_tokens_seen": 171696, |
| "step": 275 |
| }, |
| { |
| "epoch": 4.912280701754386, |
| "grad_norm": 2.9338085651397705, |
| "learning_rate": 4.6876617036064844e-05, |
| "loss": 0.3706, |
| "num_input_tokens_seen": 174672, |
| "step": 280 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.014911307021975517, |
| "learning_rate": 4.668880830394093e-05, |
| "loss": 0.1807, |
| "num_input_tokens_seen": 177224, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.14759516716003418, |
| "eval_runtime": 0.6618, |
| "eval_samples_per_second": 37.775, |
| "eval_steps_per_second": 10.577, |
| "num_input_tokens_seen": 177224, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.087719298245614, |
| "grad_norm": 0.6438454389572144, |
| "learning_rate": 4.649591596164778e-05, |
| "loss": 0.0869, |
| "num_input_tokens_seen": 180840, |
| "step": 290 |
| }, |
| { |
| "epoch": 5.175438596491228, |
| "grad_norm": 0.526551365852356, |
| "learning_rate": 4.629798522095818e-05, |
| "loss": 0.0851, |
| "num_input_tokens_seen": 183976, |
| "step": 295 |
| }, |
| { |
| "epoch": 5.2631578947368425, |
| "grad_norm": 0.8825740218162537, |
| "learning_rate": 4.6095062474588225e-05, |
| "loss": 0.2137, |
| "num_input_tokens_seen": 187016, |
| "step": 300 |
| }, |
| { |
| "epoch": 5.350877192982456, |
| "grad_norm": 1.882129430770874, |
| "learning_rate": 4.588719528532342e-05, |
| "loss": 0.1867, |
| "num_input_tokens_seen": 189480, |
| "step": 305 |
| }, |
| { |
| "epoch": 5.43859649122807, |
| "grad_norm": 0.5353156924247742, |
| "learning_rate": 4.5674432374870455e-05, |
| "loss": 0.2414, |
| "num_input_tokens_seen": 192712, |
| "step": 310 |
| }, |
| { |
| "epoch": 5.526315789473684, |
| "grad_norm": 0.5105774998664856, |
| "learning_rate": 4.545682361243748e-05, |
| "loss": 0.3162, |
| "num_input_tokens_seen": 195496, |
| "step": 315 |
| }, |
| { |
| "epoch": 5.614035087719298, |
| "grad_norm": 0.19848407804965973, |
| "learning_rate": 4.5234420003045236e-05, |
| "loss": 0.0752, |
| "num_input_tokens_seen": 199240, |
| "step": 320 |
| }, |
| { |
| "epoch": 5.701754385964913, |
| "grad_norm": 0.9869743585586548, |
| "learning_rate": 4.5007273675572104e-05, |
| "loss": 0.1525, |
| "num_input_tokens_seen": 201832, |
| "step": 325 |
| }, |
| { |
| "epoch": 5.7894736842105265, |
| "grad_norm": 0.23039640486240387, |
| "learning_rate": 4.4775437870535685e-05, |
| "loss": 0.0352, |
| "num_input_tokens_seen": 204968, |
| "step": 330 |
| }, |
| { |
| "epoch": 5.87719298245614, |
| "grad_norm": 1.3898390531539917, |
| "learning_rate": 4.4538966927613836e-05, |
| "loss": 0.0968, |
| "num_input_tokens_seen": 208360, |
| "step": 335 |
| }, |
| { |
| "epoch": 5.964912280701754, |
| "grad_norm": 0.6159465312957764, |
| "learning_rate": 4.4297916272908024e-05, |
| "loss": 0.0604, |
| "num_input_tokens_seen": 211080, |
| "step": 340 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.14964094758033752, |
| "eval_runtime": 0.6625, |
| "eval_samples_per_second": 37.738, |
| "eval_steps_per_second": 10.567, |
| "num_input_tokens_seen": 212000, |
| "step": 342 |
| }, |
| { |
| "epoch": 6.052631578947368, |
| "grad_norm": 0.6027920842170715, |
| "learning_rate": 4.405234240595214e-05, |
| "loss": 0.1657, |
| "num_input_tokens_seen": 214176, |
| "step": 345 |
| }, |
| { |
| "epoch": 6.140350877192983, |
| "grad_norm": 0.5913047790527344, |
| "learning_rate": 4.3802302886469606e-05, |
| "loss": 0.1199, |
| "num_input_tokens_seen": 216736, |
| "step": 350 |
| }, |
| { |
| "epoch": 6.228070175438597, |
| "grad_norm": 0.8113365769386292, |
| "learning_rate": 4.3547856320882044e-05, |
| "loss": 0.1447, |
| "num_input_tokens_seen": 220224, |
| "step": 355 |
| }, |
| { |
| "epoch": 6.315789473684211, |
| "grad_norm": 0.7925099730491638, |
| "learning_rate": 4.328906234857259e-05, |
| "loss": 0.062, |
| "num_input_tokens_seen": 223136, |
| "step": 360 |
| }, |
| { |
| "epoch": 6.4035087719298245, |
| "grad_norm": 1.277872920036316, |
| "learning_rate": 4.302598162790712e-05, |
| "loss": 0.1789, |
| "num_input_tokens_seen": 226944, |
| "step": 365 |
| }, |
| { |
| "epoch": 6.491228070175438, |
| "grad_norm": 0.2423498034477234, |
| "learning_rate": 4.27586758220166e-05, |
| "loss": 0.0522, |
| "num_input_tokens_seen": 230368, |
| "step": 370 |
| }, |
| { |
| "epoch": 6.578947368421053, |
| "grad_norm": 3.049635887145996, |
| "learning_rate": 4.2487207584343955e-05, |
| "loss": 0.2337, |
| "num_input_tokens_seen": 233376, |
| "step": 375 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.3379043638706207, |
| "learning_rate": 4.2211640543958796e-05, |
| "loss": 0.116, |
| "num_input_tokens_seen": 236384, |
| "step": 380 |
| }, |
| { |
| "epoch": 6.754385964912281, |
| "grad_norm": 1.3031666278839111, |
| "learning_rate": 4.193203929064353e-05, |
| "loss": 0.1001, |
| "num_input_tokens_seen": 239584, |
| "step": 385 |
| }, |
| { |
| "epoch": 6.842105263157895, |
| "grad_norm": 0.3099721372127533, |
| "learning_rate": 4.164846935975421e-05, |
| "loss": 0.241, |
| "num_input_tokens_seen": 242624, |
| "step": 390 |
| }, |
| { |
| "epoch": 6.9298245614035086, |
| "grad_norm": 0.2124738097190857, |
| "learning_rate": 4.136099721685983e-05, |
| "loss": 0.0518, |
| "num_input_tokens_seen": 246304, |
| "step": 395 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.14769574999809265, |
| "eval_runtime": 0.6619, |
| "eval_samples_per_second": 37.768, |
| "eval_steps_per_second": 10.575, |
| "num_input_tokens_seen": 248272, |
| "step": 399 |
| }, |
| { |
| "epoch": 7.017543859649122, |
| "grad_norm": 0.7997441291809082, |
| "learning_rate": 4.1069690242163484e-05, |
| "loss": 0.1552, |
| "num_input_tokens_seen": 248912, |
| "step": 400 |
| }, |
| { |
| "epoch": 7.105263157894737, |
| "grad_norm": 0.6180697083473206, |
| "learning_rate": 4.0774616714709316e-05, |
| "loss": 0.1963, |
| "num_input_tokens_seen": 252016, |
| "step": 405 |
| }, |
| { |
| "epoch": 7.192982456140351, |
| "grad_norm": 0.5003829598426819, |
| "learning_rate": 4.047584579637857e-05, |
| "loss": 0.0738, |
| "num_input_tokens_seen": 255856, |
| "step": 410 |
| }, |
| { |
| "epoch": 7.280701754385965, |
| "grad_norm": 1.1428700685501099, |
| "learning_rate": 4.0173447515678916e-05, |
| "loss": 0.2976, |
| "num_input_tokens_seen": 258800, |
| "step": 415 |
| }, |
| { |
| "epoch": 7.368421052631579, |
| "grad_norm": 0.221315398812294, |
| "learning_rate": 3.986749275133057e-05, |
| "loss": 0.0607, |
| "num_input_tokens_seen": 261584, |
| "step": 420 |
| }, |
| { |
| "epoch": 7.456140350877193, |
| "grad_norm": 0.31321048736572266, |
| "learning_rate": 3.955805321565304e-05, |
| "loss": 0.0776, |
| "num_input_tokens_seen": 265168, |
| "step": 425 |
| }, |
| { |
| "epoch": 7.543859649122807, |
| "grad_norm": 0.10731899738311768, |
| "learning_rate": 3.9245201437756654e-05, |
| "loss": 0.1641, |
| "num_input_tokens_seen": 268240, |
| "step": 430 |
| }, |
| { |
| "epoch": 7.631578947368421, |
| "grad_norm": 1.1928623914718628, |
| "learning_rate": 3.892901074654255e-05, |
| "loss": 0.127, |
| "num_input_tokens_seen": 271728, |
| "step": 435 |
| }, |
| { |
| "epoch": 7.719298245614035, |
| "grad_norm": 0.588262140750885, |
| "learning_rate": 3.860955525351516e-05, |
| "loss": 0.1401, |
| "num_input_tokens_seen": 274736, |
| "step": 440 |
| }, |
| { |
| "epoch": 7.807017543859649, |
| "grad_norm": 0.5087107419967651, |
| "learning_rate": 3.82869098354114e-05, |
| "loss": 0.1377, |
| "num_input_tokens_seen": 278096, |
| "step": 445 |
| }, |
| { |
| "epoch": 7.894736842105263, |
| "grad_norm": 0.40772467851638794, |
| "learning_rate": 3.796115011665034e-05, |
| "loss": 0.0619, |
| "num_input_tokens_seen": 280592, |
| "step": 450 |
| }, |
| { |
| "epoch": 7.982456140350877, |
| "grad_norm": 1.2497111558914185, |
| "learning_rate": 3.763235245160775e-05, |
| "loss": 0.0906, |
| "num_input_tokens_seen": 284144, |
| "step": 455 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.14076876640319824, |
| "eval_runtime": 0.659, |
| "eval_samples_per_second": 37.938, |
| "eval_steps_per_second": 10.623, |
| "num_input_tokens_seen": 284248, |
| "step": 456 |
| }, |
| { |
| "epoch": 8.070175438596491, |
| "grad_norm": 0.9935876131057739, |
| "learning_rate": 3.7300593906719464e-05, |
| "loss": 0.1387, |
| "num_input_tokens_seen": 287128, |
| "step": 460 |
| }, |
| { |
| "epoch": 8.157894736842104, |
| "grad_norm": 0.37087738513946533, |
| "learning_rate": 3.69659522424179e-05, |
| "loss": 0.0439, |
| "num_input_tokens_seen": 290744, |
| "step": 465 |
| }, |
| { |
| "epoch": 8.24561403508772, |
| "grad_norm": 0.5564098358154297, |
| "learning_rate": 3.662850589490592e-05, |
| "loss": 0.0702, |
| "num_input_tokens_seen": 293816, |
| "step": 470 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.8945322632789612, |
| "learning_rate": 3.628833395777224e-05, |
| "loss": 0.2406, |
| "num_input_tokens_seen": 297080, |
| "step": 475 |
| }, |
| { |
| "epoch": 8.421052631578947, |
| "grad_norm": 0.9603311419487, |
| "learning_rate": 3.59455161634528e-05, |
| "loss": 0.1544, |
| "num_input_tokens_seen": 300408, |
| "step": 480 |
| }, |
| { |
| "epoch": 8.508771929824562, |
| "grad_norm": 1.0279154777526855, |
| "learning_rate": 3.560013286454242e-05, |
| "loss": 0.1642, |
| "num_input_tokens_seen": 303384, |
| "step": 485 |
| }, |
| { |
| "epoch": 8.596491228070175, |
| "grad_norm": 3.1768276691436768, |
| "learning_rate": 3.5252265014961006e-05, |
| "loss": 0.1795, |
| "num_input_tokens_seen": 306232, |
| "step": 490 |
| }, |
| { |
| "epoch": 8.68421052631579, |
| "grad_norm": 0.07754722237586975, |
| "learning_rate": 3.490199415097892e-05, |
| "loss": 0.0485, |
| "num_input_tokens_seen": 309816, |
| "step": 495 |
| }, |
| { |
| "epoch": 8.771929824561404, |
| "grad_norm": 0.3435470759868622, |
| "learning_rate": 3.45494023721058e-05, |
| "loss": 0.0644, |
| "num_input_tokens_seen": 312408, |
| "step": 500 |
| }, |
| { |
| "epoch": 8.859649122807017, |
| "grad_norm": 0.9218394160270691, |
| "learning_rate": 3.4194572321847336e-05, |
| "loss": 0.1169, |
| "num_input_tokens_seen": 315416, |
| "step": 505 |
| }, |
| { |
| "epoch": 8.947368421052632, |
| "grad_norm": 0.2821664810180664, |
| "learning_rate": 3.383758716833459e-05, |
| "loss": 0.1919, |
| "num_input_tokens_seen": 318392, |
| "step": 510 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.13670699298381805, |
| "eval_runtime": 0.6656, |
| "eval_samples_per_second": 37.559, |
| "eval_steps_per_second": 10.517, |
| "num_input_tokens_seen": 319488, |
| "step": 513 |
| }, |
| { |
| "epoch": 9.035087719298245, |
| "grad_norm": 1.047012209892273, |
| "learning_rate": 3.347853058483037e-05, |
| "loss": 0.078, |
| "num_input_tokens_seen": 320832, |
| "step": 515 |
| }, |
| { |
| "epoch": 9.12280701754386, |
| "grad_norm": 0.3289402425289154, |
| "learning_rate": 3.311748673011709e-05, |
| "loss": 0.0758, |
| "num_input_tokens_seen": 323840, |
| "step": 520 |
| }, |
| { |
| "epoch": 9.210526315789474, |
| "grad_norm": 0.8709810972213745, |
| "learning_rate": 3.275454022877097e-05, |
| "loss": 0.0834, |
| "num_input_tokens_seen": 327200, |
| "step": 525 |
| }, |
| { |
| "epoch": 9.298245614035087, |
| "grad_norm": 1.6551891565322876, |
| "learning_rate": 3.238977615132697e-05, |
| "loss": 0.0632, |
| "num_input_tokens_seen": 330208, |
| "step": 530 |
| }, |
| { |
| "epoch": 9.385964912280702, |
| "grad_norm": 0.17781463265419006, |
| "learning_rate": 3.202327999433924e-05, |
| "loss": 0.043, |
| "num_input_tokens_seen": 332864, |
| "step": 535 |
| }, |
| { |
| "epoch": 9.473684210526315, |
| "grad_norm": 0.15141192078590393, |
| "learning_rate": 3.165513766034167e-05, |
| "loss": 0.2576, |
| "num_input_tokens_seen": 336224, |
| "step": 540 |
| }, |
| { |
| "epoch": 9.56140350877193, |
| "grad_norm": 0.87317955493927, |
| "learning_rate": 3.128543543771336e-05, |
| "loss": 0.0637, |
| "num_input_tokens_seen": 339392, |
| "step": 545 |
| }, |
| { |
| "epoch": 9.649122807017545, |
| "grad_norm": 0.5003905892372131, |
| "learning_rate": 3.091425998045356e-05, |
| "loss": 0.088, |
| "num_input_tokens_seen": 342624, |
| "step": 550 |
| }, |
| { |
| "epoch": 9.736842105263158, |
| "grad_norm": 2.310898542404175, |
| "learning_rate": 3.0541698287870965e-05, |
| "loss": 0.1858, |
| "num_input_tokens_seen": 345472, |
| "step": 555 |
| }, |
| { |
| "epoch": 9.824561403508772, |
| "grad_norm": 0.5038716793060303, |
| "learning_rate": 3.01678376841921e-05, |
| "loss": 0.2795, |
| "num_input_tokens_seen": 348160, |
| "step": 560 |
| }, |
| { |
| "epoch": 9.912280701754385, |
| "grad_norm": 0.5883809924125671, |
| "learning_rate": 2.9792765798093465e-05, |
| "loss": 0.1517, |
| "num_input_tokens_seen": 351456, |
| "step": 565 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.7664790153503418, |
| "learning_rate": 2.94165705421624e-05, |
| "loss": 0.0829, |
| "num_input_tokens_seen": 354472, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.14112165570259094, |
| "eval_runtime": 0.66, |
| "eval_samples_per_second": 37.877, |
| "eval_steps_per_second": 10.606, |
| "num_input_tokens_seen": 354472, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.087719298245615, |
| "grad_norm": 0.09412403404712677, |
| "learning_rate": 2.9039340092291373e-05, |
| "loss": 0.0643, |
| "num_input_tokens_seen": 358024, |
| "step": 575 |
| }, |
| { |
| "epoch": 10.175438596491228, |
| "grad_norm": 0.3542427122592926, |
| "learning_rate": 2.8661162867010543e-05, |
| "loss": 0.1596, |
| "num_input_tokens_seen": 361192, |
| "step": 580 |
| }, |
| { |
| "epoch": 10.263157894736842, |
| "grad_norm": 0.6384028196334839, |
| "learning_rate": 2.8282127506763456e-05, |
| "loss": 0.094, |
| "num_input_tokens_seen": 363816, |
| "step": 585 |
| }, |
| { |
| "epoch": 10.350877192982455, |
| "grad_norm": 0.14515523612499237, |
| "learning_rate": 2.7902322853130757e-05, |
| "loss": 0.168, |
| "num_input_tokens_seen": 366440, |
| "step": 590 |
| }, |
| { |
| "epoch": 10.43859649122807, |
| "grad_norm": 0.7717634439468384, |
| "learning_rate": 2.752183792800671e-05, |
| "loss": 0.0475, |
| "num_input_tokens_seen": 369672, |
| "step": 595 |
| }, |
| { |
| "epoch": 10.526315789473685, |
| "grad_norm": 0.3233764171600342, |
| "learning_rate": 2.7140761912733474e-05, |
| "loss": 0.0767, |
| "num_input_tokens_seen": 372808, |
| "step": 600 |
| }, |
| { |
| "epoch": 10.614035087719298, |
| "grad_norm": 0.7295190691947937, |
| "learning_rate": 2.6759184127198046e-05, |
| "loss": 0.232, |
| "num_input_tokens_seen": 375560, |
| "step": 605 |
| }, |
| { |
| "epoch": 10.701754385964913, |
| "grad_norm": 0.40313610434532166, |
| "learning_rate": 2.6377194008896637e-05, |
| "loss": 0.0509, |
| "num_input_tokens_seen": 378792, |
| "step": 610 |
| }, |
| { |
| "epoch": 10.789473684210526, |
| "grad_norm": 0.6811980605125427, |
| "learning_rate": 2.5994881091971605e-05, |
| "loss": 0.1213, |
| "num_input_tokens_seen": 382312, |
| "step": 615 |
| }, |
| { |
| "epoch": 10.87719298245614, |
| "grad_norm": 1.1628668308258057, |
| "learning_rate": 2.5612334986225623e-05, |
| "loss": 0.1247, |
| "num_input_tokens_seen": 386088, |
| "step": 620 |
| }, |
| { |
| "epoch": 10.964912280701755, |
| "grad_norm": 0.5735217332839966, |
| "learning_rate": 2.5229645356118163e-05, |
| "loss": 0.0778, |
| "num_input_tokens_seen": 388840, |
| "step": 625 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 0.14512333273887634, |
| "eval_runtime": 0.6618, |
| "eval_samples_per_second": 37.774, |
| "eval_steps_per_second": 10.577, |
| "num_input_tokens_seen": 389408, |
| "step": 627 |
| }, |
| { |
| "epoch": 11.052631578947368, |
| "grad_norm": 1.5689105987548828, |
| "learning_rate": 2.4846901899749185e-05, |
| "loss": 0.1349, |
| "num_input_tokens_seen": 391200, |
| "step": 630 |
| }, |
| { |
| "epoch": 11.140350877192983, |
| "grad_norm": 1.2222450971603394, |
| "learning_rate": 2.4464194327834926e-05, |
| "loss": 0.2667, |
| "num_input_tokens_seen": 394880, |
| "step": 635 |
| }, |
| { |
| "epoch": 11.228070175438596, |
| "grad_norm": 0.22504912316799164, |
| "learning_rate": 2.4081612342680694e-05, |
| "loss": 0.0389, |
| "num_input_tokens_seen": 398336, |
| "step": 640 |
| }, |
| { |
| "epoch": 11.31578947368421, |
| "grad_norm": 0.16236864030361176, |
| "learning_rate": 2.369924561715569e-05, |
| "loss": 0.0645, |
| "num_input_tokens_seen": 401632, |
| "step": 645 |
| }, |
| { |
| "epoch": 11.403508771929825, |
| "grad_norm": 0.9279342889785767, |
| "learning_rate": 2.3317183773674718e-05, |
| "loss": 0.0926, |
| "num_input_tokens_seen": 405120, |
| "step": 650 |
| }, |
| { |
| "epoch": 11.491228070175438, |
| "grad_norm": 0.2551843822002411, |
| "learning_rate": 2.2935516363191693e-05, |
| "loss": 0.285, |
| "num_input_tokens_seen": 408192, |
| "step": 655 |
| }, |
| { |
| "epoch": 11.578947368421053, |
| "grad_norm": 0.9922687411308289, |
| "learning_rate": 2.2554332844209904e-05, |
| "loss": 0.0657, |
| "num_input_tokens_seen": 411360, |
| "step": 660 |
| }, |
| { |
| "epoch": 11.666666666666666, |
| "grad_norm": 0.32865116000175476, |
| "learning_rate": 2.2173722561813987e-05, |
| "loss": 0.0993, |
| "num_input_tokens_seen": 414144, |
| "step": 665 |
| }, |
| { |
| "epoch": 11.75438596491228, |
| "grad_norm": 0.1407676786184311, |
| "learning_rate": 2.179377472672842e-05, |
| "loss": 0.1065, |
| "num_input_tokens_seen": 417760, |
| "step": 670 |
| }, |
| { |
| "epoch": 11.842105263157894, |
| "grad_norm": 0.33401456475257874, |
| "learning_rate": 2.1414578394407597e-05, |
| "loss": 0.1535, |
| "num_input_tokens_seen": 420640, |
| "step": 675 |
| }, |
| { |
| "epoch": 11.929824561403509, |
| "grad_norm": 0.7902398705482483, |
| "learning_rate": 2.1036222444162147e-05, |
| "loss": 0.1052, |
| "num_input_tokens_seen": 423360, |
| "step": 680 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 0.14079952239990234, |
| "eval_runtime": 0.6598, |
| "eval_samples_per_second": 37.889, |
| "eval_steps_per_second": 10.609, |
| "num_input_tokens_seen": 425328, |
| "step": 684 |
| }, |
| { |
| "epoch": 12.017543859649123, |
| "grad_norm": 0.6184465885162354, |
| "learning_rate": 2.0658795558326743e-05, |
| "loss": 0.0427, |
| "num_input_tokens_seen": 426288, |
| "step": 685 |
| }, |
| { |
| "epoch": 12.105263157894736, |
| "grad_norm": 0.45558130741119385, |
| "learning_rate": 2.0282386201473894e-05, |
| "loss": 0.1015, |
| "num_input_tokens_seen": 429136, |
| "step": 690 |
| }, |
| { |
| "epoch": 12.192982456140351, |
| "grad_norm": 0.3064878284931183, |
| "learning_rate": 1.99070825996789e-05, |
| "loss": 0.1169, |
| "num_input_tokens_seen": 432272, |
| "step": 695 |
| }, |
| { |
| "epoch": 12.280701754385966, |
| "grad_norm": 0.8751903176307678, |
| "learning_rate": 1.9532972719840607e-05, |
| "loss": 0.0688, |
| "num_input_tokens_seen": 435760, |
| "step": 700 |
| }, |
| { |
| "epoch": 12.368421052631579, |
| "grad_norm": 0.65683513879776, |
| "learning_rate": 1.9160144249063035e-05, |
| "loss": 0.1505, |
| "num_input_tokens_seen": 439312, |
| "step": 705 |
| }, |
| { |
| "epoch": 12.456140350877194, |
| "grad_norm": 0.7957984209060669, |
| "learning_rate": 1.8788684574102467e-05, |
| "loss": 0.044, |
| "num_input_tokens_seen": 442640, |
| "step": 710 |
| }, |
| { |
| "epoch": 12.543859649122806, |
| "grad_norm": 0.8334346413612366, |
| "learning_rate": 1.8418680760885027e-05, |
| "loss": 0.1376, |
| "num_input_tokens_seen": 445648, |
| "step": 715 |
| }, |
| { |
| "epoch": 12.631578947368421, |
| "grad_norm": 0.556850016117096, |
| "learning_rate": 1.805021953409934e-05, |
| "loss": 0.2319, |
| "num_input_tokens_seen": 448208, |
| "step": 720 |
| }, |
| { |
| "epoch": 12.719298245614034, |
| "grad_norm": 0.9198157787322998, |
| "learning_rate": 1.7683387256869353e-05, |
| "loss": 0.087, |
| "num_input_tokens_seen": 451600, |
| "step": 725 |
| }, |
| { |
| "epoch": 12.807017543859649, |
| "grad_norm": 0.624397337436676, |
| "learning_rate": 1.7318269910511736e-05, |
| "loss": 0.0978, |
| "num_input_tokens_seen": 454992, |
| "step": 730 |
| }, |
| { |
| "epoch": 12.894736842105264, |
| "grad_norm": 0.14907003939151764, |
| "learning_rate": 1.6954953074382863e-05, |
| "loss": 0.0331, |
| "num_input_tokens_seen": 458032, |
| "step": 735 |
| }, |
| { |
| "epoch": 12.982456140350877, |
| "grad_norm": 0.08006937801837921, |
| "learning_rate": 1.659352190581993e-05, |
| "loss": 0.2006, |
| "num_input_tokens_seen": 461104, |
| "step": 740 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_loss": 0.14212080836296082, |
| "eval_runtime": 0.6621, |
| "eval_samples_per_second": 37.761, |
| "eval_steps_per_second": 10.573, |
| "num_input_tokens_seen": 461216, |
| "step": 741 |
| }, |
| { |
| "epoch": 13.070175438596491, |
| "grad_norm": 1.6945339441299438, |
| "learning_rate": 1.6234061120181142e-05, |
| "loss": 0.1481, |
| "num_input_tokens_seen": 463904, |
| "step": 745 |
| }, |
| { |
| "epoch": 13.157894736842104, |
| "grad_norm": 0.9081370234489441, |
| "learning_rate": 1.5876654970989308e-05, |
| "loss": 0.0697, |
| "num_input_tokens_seen": 467328, |
| "step": 750 |
| }, |
| { |
| "epoch": 13.24561403508772, |
| "grad_norm": 0.7696995139122009, |
| "learning_rate": 1.552138723018382e-05, |
| "loss": 0.1633, |
| "num_input_tokens_seen": 469792, |
| "step": 755 |
| }, |
| { |
| "epoch": 13.333333333333334, |
| "grad_norm": 0.43313539028167725, |
| "learning_rate": 1.5168341168485423e-05, |
| "loss": 0.0502, |
| "num_input_tokens_seen": 472896, |
| "step": 760 |
| }, |
| { |
| "epoch": 13.421052631578947, |
| "grad_norm": 0.2849214971065521, |
| "learning_rate": 1.4817599535878565e-05, |
| "loss": 0.1108, |
| "num_input_tokens_seen": 475904, |
| "step": 765 |
| }, |
| { |
| "epoch": 13.508771929824562, |
| "grad_norm": 0.49461111426353455, |
| "learning_rate": 1.4469244542215682e-05, |
| "loss": 0.3415, |
| "num_input_tokens_seen": 478848, |
| "step": 770 |
| }, |
| { |
| "epoch": 13.596491228070175, |
| "grad_norm": 0.5328129529953003, |
| "learning_rate": 1.4123357837948175e-05, |
| "loss": 0.1263, |
| "num_input_tokens_seen": 482080, |
| "step": 775 |
| }, |
| { |
| "epoch": 13.68421052631579, |
| "grad_norm": 0.3329578936100006, |
| "learning_rate": 1.3780020494988446e-05, |
| "loss": 0.0398, |
| "num_input_tokens_seen": 485440, |
| "step": 780 |
| }, |
| { |
| "epoch": 13.771929824561404, |
| "grad_norm": 0.361813485622406, |
| "learning_rate": 1.3439312987707615e-05, |
| "loss": 0.0591, |
| "num_input_tokens_seen": 488640, |
| "step": 785 |
| }, |
| { |
| "epoch": 13.859649122807017, |
| "grad_norm": 0.16096878051757812, |
| "learning_rate": 1.3101315174073162e-05, |
| "loss": 0.0569, |
| "num_input_tokens_seen": 492032, |
| "step": 790 |
| }, |
| { |
| "epoch": 13.947368421052632, |
| "grad_norm": 0.8325091600418091, |
| "learning_rate": 1.2766106276931223e-05, |
| "loss": 0.0984, |
| "num_input_tokens_seen": 495264, |
| "step": 795 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_loss": 0.14145123958587646, |
| "eval_runtime": 0.661, |
| "eval_samples_per_second": 37.822, |
| "eval_steps_per_second": 10.59, |
| "num_input_tokens_seen": 496704, |
| "step": 798 |
| }, |
| { |
| "epoch": 14.035087719298245, |
| "grad_norm": 0.19925180077552795, |
| "learning_rate": 1.243376486543755e-05, |
| "loss": 0.0408, |
| "num_input_tokens_seen": 498464, |
| "step": 800 |
| }, |
| { |
| "epoch": 14.12280701754386, |
| "grad_norm": 1.0538097620010376, |
| "learning_rate": 1.2104368836641908e-05, |
| "loss": 0.0608, |
| "num_input_tokens_seen": 501632, |
| "step": 805 |
| }, |
| { |
| "epoch": 14.210526315789474, |
| "grad_norm": 0.5036182403564453, |
| "learning_rate": 1.1777995397229771e-05, |
| "loss": 0.0278, |
| "num_input_tokens_seen": 504544, |
| "step": 810 |
| }, |
| { |
| "epoch": 14.298245614035087, |
| "grad_norm": 0.7434744834899902, |
| "learning_rate": 1.1454721045426073e-05, |
| "loss": 0.1002, |
| "num_input_tokens_seen": 507456, |
| "step": 815 |
| }, |
| { |
| "epoch": 14.385964912280702, |
| "grad_norm": 0.37160205841064453, |
| "learning_rate": 1.113462155306478e-05, |
| "loss": 0.0348, |
| "num_input_tokens_seen": 511136, |
| "step": 820 |
| }, |
| { |
| "epoch": 14.473684210526315, |
| "grad_norm": 0.7353971004486084, |
| "learning_rate": 1.0817771947828934e-05, |
| "loss": 0.0651, |
| "num_input_tokens_seen": 514368, |
| "step": 825 |
| }, |
| { |
| "epoch": 14.56140350877193, |
| "grad_norm": 0.6230506300926208, |
| "learning_rate": 1.0504246495664932e-05, |
| "loss": 0.1953, |
| "num_input_tokens_seen": 517248, |
| "step": 830 |
| }, |
| { |
| "epoch": 14.649122807017545, |
| "grad_norm": 0.5421151518821716, |
| "learning_rate": 1.0194118683375503e-05, |
| "loss": 0.3345, |
| "num_input_tokens_seen": 520160, |
| "step": 835 |
| }, |
| { |
| "epoch": 14.736842105263158, |
| "grad_norm": 1.370159387588501, |
| "learning_rate": 9.887461201395176e-06, |
| "loss": 0.0612, |
| "num_input_tokens_seen": 523424, |
| "step": 840 |
| }, |
| { |
| "epoch": 14.824561403508772, |
| "grad_norm": 0.9853245615959167, |
| "learning_rate": 9.584345926752524e-06, |
| "loss": 0.166, |
| "num_input_tokens_seen": 526240, |
| "step": 845 |
| }, |
| { |
| "epoch": 14.912280701754385, |
| "grad_norm": 0.9621556401252747, |
| "learning_rate": 9.284843906222948e-06, |
| "loss": 0.1395, |
| "num_input_tokens_seen": 529440, |
| "step": 850 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 0.4544803500175476, |
| "learning_rate": 8.98902533967618e-06, |
| "loss": 0.0945, |
| "num_input_tokens_seen": 532504, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.140976220369339, |
| "eval_runtime": 0.6619, |
| "eval_samples_per_second": 37.77, |
| "eval_steps_per_second": 10.576, |
| "num_input_tokens_seen": 532504, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.087719298245615, |
| "grad_norm": 0.6265756487846375, |
| "learning_rate": 8.696959563622174e-06, |
| "loss": 0.0319, |
| "num_input_tokens_seen": 535544, |
| "step": 860 |
| }, |
| { |
| "epoch": 15.175438596491228, |
| "grad_norm": 0.9495764970779419, |
| "learning_rate": 8.40871503495947e-06, |
| "loss": 0.0901, |
| "num_input_tokens_seen": 538744, |
| "step": 865 |
| }, |
| { |
| "epoch": 15.263157894736842, |
| "grad_norm": 0.44339001178741455, |
| "learning_rate": 8.124359314929622e-06, |
| "loss": 0.0529, |
| "num_input_tokens_seen": 542040, |
| "step": 870 |
| }, |
| { |
| "epoch": 15.350877192982455, |
| "grad_norm": 0.3083951771259308, |
| "learning_rate": 7.843959053281663e-06, |
| "loss": 0.0759, |
| "num_input_tokens_seen": 544888, |
| "step": 875 |
| }, |
| { |
| "epoch": 15.43859649122807, |
| "grad_norm": 2.1893868446350098, |
| "learning_rate": 7.5675799726501155e-06, |
| "loss": 0.2418, |
| "num_input_tokens_seen": 547640, |
| "step": 880 |
| }, |
| { |
| "epoch": 15.526315789473685, |
| "grad_norm": 0.17266622185707092, |
| "learning_rate": 7.295286853150391e-06, |
| "loss": 0.245, |
| "num_input_tokens_seen": 551224, |
| "step": 885 |
| }, |
| { |
| "epoch": 15.614035087719298, |
| "grad_norm": 0.4342977702617645, |
| "learning_rate": 7.027143517195023e-06, |
| "loss": 0.0359, |
| "num_input_tokens_seen": 554264, |
| "step": 890 |
| }, |
| { |
| "epoch": 15.701754385964913, |
| "grad_norm": 0.06123614311218262, |
| "learning_rate": 6.763212814534484e-06, |
| "loss": 0.12, |
| "num_input_tokens_seen": 557880, |
| "step": 895 |
| }, |
| { |
| "epoch": 15.789473684210526, |
| "grad_norm": 0.9682021737098694, |
| "learning_rate": 6.503556607525838e-06, |
| "loss": 0.0877, |
| "num_input_tokens_seen": 560376, |
| "step": 900 |
| }, |
| { |
| "epoch": 15.87719298245614, |
| "grad_norm": 1.33763587474823, |
| "learning_rate": 6.248235756632984e-06, |
| "loss": 0.0926, |
| "num_input_tokens_seen": 563864, |
| "step": 905 |
| }, |
| { |
| "epoch": 15.964912280701755, |
| "grad_norm": 0.6766443848609924, |
| "learning_rate": 5.997310106161589e-06, |
| "loss": 0.0651, |
| "num_input_tokens_seen": 567352, |
| "step": 910 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_loss": 0.14221011102199554, |
| "eval_runtime": 0.6615, |
| "eval_samples_per_second": 37.796, |
| "eval_steps_per_second": 10.583, |
| "num_input_tokens_seen": 567952, |
| "step": 912 |
| }, |
| { |
| "epoch": 16.05263157894737, |
| "grad_norm": 3.150055408477783, |
| "learning_rate": 5.7508384702323226e-06, |
| "loss": 0.3432, |
| "num_input_tokens_seen": 570448, |
| "step": 915 |
| }, |
| { |
| "epoch": 16.140350877192983, |
| "grad_norm": 0.22570478916168213, |
| "learning_rate": 5.508878618995439e-06, |
| "loss": 0.1023, |
| "num_input_tokens_seen": 573808, |
| "step": 920 |
| }, |
| { |
| "epoch": 16.228070175438596, |
| "grad_norm": 0.14946742355823517, |
| "learning_rate": 5.271487265090163e-06, |
| "loss": 0.053, |
| "num_input_tokens_seen": 577200, |
| "step": 925 |
| }, |
| { |
| "epoch": 16.31578947368421, |
| "grad_norm": 0.17775960266590118, |
| "learning_rate": 5.038720050351842e-06, |
| "loss": 0.0826, |
| "num_input_tokens_seen": 580240, |
| "step": 930 |
| }, |
| { |
| "epoch": 16.403508771929825, |
| "grad_norm": 0.5061619877815247, |
| "learning_rate": 4.810631532770182e-06, |
| "loss": 0.095, |
| "num_input_tokens_seen": 583120, |
| "step": 935 |
| }, |
| { |
| "epoch": 16.49122807017544, |
| "grad_norm": 0.6057607531547546, |
| "learning_rate": 4.587275173701428e-06, |
| "loss": 0.02, |
| "num_input_tokens_seen": 586032, |
| "step": 940 |
| }, |
| { |
| "epoch": 16.57894736842105, |
| "grad_norm": 0.918347179889679, |
| "learning_rate": 4.368703325337667e-06, |
| "loss": 0.0397, |
| "num_input_tokens_seen": 589392, |
| "step": 945 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 1.3691737651824951, |
| "learning_rate": 4.154967218436037e-06, |
| "loss": 0.1686, |
| "num_input_tokens_seen": 592624, |
| "step": 950 |
| }, |
| { |
| "epoch": 16.75438596491228, |
| "grad_norm": 0.5735297203063965, |
| "learning_rate": 3.94611695031086e-06, |
| "loss": 0.1428, |
| "num_input_tokens_seen": 595440, |
| "step": 955 |
| }, |
| { |
| "epoch": 16.842105263157894, |
| "grad_norm": 2.792652130126953, |
| "learning_rate": 3.74220147309135e-06, |
| "loss": 0.1857, |
| "num_input_tokens_seen": 598640, |
| "step": 960 |
| }, |
| { |
| "epoch": 16.92982456140351, |
| "grad_norm": 1.5320689678192139, |
| "learning_rate": 3.543268582247844e-06, |
| "loss": 0.101, |
| "num_input_tokens_seen": 601584, |
| "step": 965 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_loss": 0.14109651744365692, |
| "eval_runtime": 0.6617, |
| "eval_samples_per_second": 37.784, |
| "eval_steps_per_second": 10.58, |
| "num_input_tokens_seen": 603760, |
| "step": 969 |
| }, |
| { |
| "epoch": 17.017543859649123, |
| "grad_norm": 0.30703824758529663, |
| "learning_rate": 3.3493649053890326e-06, |
| "loss": 0.036, |
| "num_input_tokens_seen": 604464, |
| "step": 970 |
| }, |
| { |
| "epoch": 17.105263157894736, |
| "grad_norm": 0.18359772861003876, |
| "learning_rate": 3.1605358913330385e-06, |
| "loss": 0.0181, |
| "num_input_tokens_seen": 607600, |
| "step": 975 |
| }, |
| { |
| "epoch": 17.19298245614035, |
| "grad_norm": 0.19343017041683197, |
| "learning_rate": 2.9768257994546662e-06, |
| "loss": 0.0422, |
| "num_input_tokens_seen": 611152, |
| "step": 980 |
| }, |
| { |
| "epoch": 17.280701754385966, |
| "grad_norm": 0.3020448684692383, |
| "learning_rate": 2.7982776893115627e-06, |
| "loss": 0.0947, |
| "num_input_tokens_seen": 614768, |
| "step": 985 |
| }, |
| { |
| "epoch": 17.36842105263158, |
| "grad_norm": 0.04975567013025284, |
| "learning_rate": 2.624933410551508e-06, |
| "loss": 0.0781, |
| "num_input_tokens_seen": 618384, |
| "step": 990 |
| }, |
| { |
| "epoch": 17.45614035087719, |
| "grad_norm": 0.23732353746891022, |
| "learning_rate": 2.456833593103361e-06, |
| "loss": 0.1076, |
| "num_input_tokens_seen": 621424, |
| "step": 995 |
| }, |
| { |
| "epoch": 17.54385964912281, |
| "grad_norm": 0.5540131330490112, |
| "learning_rate": 2.2940176376538445e-06, |
| "loss": 0.0784, |
| "num_input_tokens_seen": 624464, |
| "step": 1000 |
| }, |
| { |
| "epoch": 17.63157894736842, |
| "grad_norm": 0.8552115559577942, |
| "learning_rate": 2.136523706412477e-06, |
| "loss": 0.0943, |
| "num_input_tokens_seen": 627824, |
| "step": 1005 |
| }, |
| { |
| "epoch": 17.719298245614034, |
| "grad_norm": 3.2258503437042236, |
| "learning_rate": 1.984388714166799e-06, |
| "loss": 0.2444, |
| "num_input_tokens_seen": 631056, |
| "step": 1010 |
| }, |
| { |
| "epoch": 17.80701754385965, |
| "grad_norm": 0.531158447265625, |
| "learning_rate": 1.837648319629956e-06, |
| "loss": 0.1473, |
| "num_input_tokens_seen": 634128, |
| "step": 1015 |
| }, |
| { |
| "epoch": 17.894736842105264, |
| "grad_norm": 1.2506383657455444, |
| "learning_rate": 1.6963369170826943e-06, |
| "loss": 0.1272, |
| "num_input_tokens_seen": 636656, |
| "step": 1020 |
| }, |
| { |
| "epoch": 17.982456140350877, |
| "grad_norm": 0.9487596154212952, |
| "learning_rate": 1.5604876283117326e-06, |
| "loss": 0.1725, |
| "num_input_tokens_seen": 639696, |
| "step": 1025 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_loss": 0.1413334310054779, |
| "eval_runtime": 0.6606, |
| "eval_samples_per_second": 37.843, |
| "eval_steps_per_second": 10.596, |
| "num_input_tokens_seen": 639784, |
| "step": 1026 |
| }, |
| { |
| "epoch": 18.07017543859649, |
| "grad_norm": 0.37698811292648315, |
| "learning_rate": 1.4301322948464147e-06, |
| "loss": 0.0566, |
| "num_input_tokens_seen": 642280, |
| "step": 1030 |
| }, |
| { |
| "epoch": 18.157894736842106, |
| "grad_norm": 0.1610272377729416, |
| "learning_rate": 1.3053014704953987e-06, |
| "loss": 0.0411, |
| "num_input_tokens_seen": 645128, |
| "step": 1035 |
| }, |
| { |
| "epoch": 18.24561403508772, |
| "grad_norm": 0.4164300858974457, |
| "learning_rate": 1.1860244141851773e-06, |
| "loss": 0.0546, |
| "num_input_tokens_seen": 647528, |
| "step": 1040 |
| }, |
| { |
| "epoch": 18.333333333333332, |
| "grad_norm": 0.30089327692985535, |
| "learning_rate": 1.0723290831021471e-06, |
| "loss": 0.0774, |
| "num_input_tokens_seen": 651560, |
| "step": 1045 |
| }, |
| { |
| "epoch": 18.42105263157895, |
| "grad_norm": 1.157859206199646, |
| "learning_rate": 9.642421261397472e-07, |
| "loss": 0.2963, |
| "num_input_tokens_seen": 654568, |
| "step": 1050 |
| }, |
| { |
| "epoch": 18.50877192982456, |
| "grad_norm": 0.9782983660697937, |
| "learning_rate": 8.617888776522642e-07, |
| "loss": 0.1507, |
| "num_input_tokens_seen": 657448, |
| "step": 1055 |
| }, |
| { |
| "epoch": 18.596491228070175, |
| "grad_norm": 0.8035924434661865, |
| "learning_rate": 7.649933515167407e-07, |
| "loss": 0.0944, |
| "num_input_tokens_seen": 660392, |
| "step": 1060 |
| }, |
| { |
| "epoch": 18.68421052631579, |
| "grad_norm": 0.5728412866592407, |
| "learning_rate": 6.738782355044049e-07, |
| "loss": 0.1625, |
| "num_input_tokens_seen": 663656, |
| "step": 1065 |
| }, |
| { |
| "epoch": 18.771929824561404, |
| "grad_norm": 0.27566325664520264, |
| "learning_rate": 5.88464885962911e-07, |
| "loss": 0.0847, |
| "num_input_tokens_seen": 667208, |
| "step": 1070 |
| }, |
| { |
| "epoch": 18.859649122807017, |
| "grad_norm": 0.47747474908828735, |
| "learning_rate": 5.087733228106517e-07, |
| "loss": 0.0614, |
| "num_input_tokens_seen": 670568, |
| "step": 1075 |
| }, |
| { |
| "epoch": 18.94736842105263, |
| "grad_norm": 0.9515513181686401, |
| "learning_rate": 4.3482222484432513e-07, |
| "loss": 0.1481, |
| "num_input_tokens_seen": 674024, |
| "step": 1080 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_loss": 0.14035473763942719, |
| "eval_runtime": 0.6645, |
| "eval_samples_per_second": 37.621, |
| "eval_steps_per_second": 10.534, |
| "num_input_tokens_seen": 675800, |
| "step": 1083 |
| }, |
| { |
| "epoch": 19.035087719298247, |
| "grad_norm": 2.027777910232544, |
| "learning_rate": 3.666289253608235e-07, |
| "loss": 0.1101, |
| "num_input_tokens_seen": 676920, |
| "step": 1085 |
| }, |
| { |
| "epoch": 19.12280701754386, |
| "grad_norm": 0.8347691893577576, |
| "learning_rate": 3.0420940809451624e-07, |
| "loss": 0.0711, |
| "num_input_tokens_seen": 679896, |
| "step": 1090 |
| }, |
| { |
| "epoch": 19.210526315789473, |
| "grad_norm": 0.572735071182251, |
| "learning_rate": 2.47578303470844e-07, |
| "loss": 0.0795, |
| "num_input_tokens_seen": 683288, |
| "step": 1095 |
| }, |
| { |
| "epoch": 19.29824561403509, |
| "grad_norm": 0.2518613934516907, |
| "learning_rate": 1.96748885177106e-07, |
| "loss": 0.123, |
| "num_input_tokens_seen": 686168, |
| "step": 1100 |
| }, |
| { |
| "epoch": 19.385964912280702, |
| "grad_norm": 0.49396681785583496, |
| "learning_rate": 1.517330670512629e-07, |
| "loss": 0.0288, |
| "num_input_tokens_seen": 689016, |
| "step": 1105 |
| }, |
| { |
| "epoch": 19.473684210526315, |
| "grad_norm": 0.5919811129570007, |
| "learning_rate": 1.125414002894759e-07, |
| "loss": 0.0933, |
| "num_input_tokens_seen": 692376, |
| "step": 1110 |
| }, |
| { |
| "epoch": 19.56140350877193, |
| "grad_norm": 0.32371166348457336, |
| "learning_rate": 7.918307097301014e-08, |
| "loss": 0.1091, |
| "num_input_tokens_seen": 695480, |
| "step": 1115 |
| }, |
| { |
| "epoch": 19.649122807017545, |
| "grad_norm": 0.7653956413269043, |
| "learning_rate": 5.166589791513465e-08, |
| "loss": 0.2382, |
| "num_input_tokens_seen": 698424, |
| "step": 1120 |
| }, |
| { |
| "epoch": 19.736842105263158, |
| "grad_norm": 0.5556171536445618, |
| "learning_rate": 2.999633082847453e-08, |
| "loss": 0.0594, |
| "num_input_tokens_seen": 701592, |
| "step": 1125 |
| }, |
| { |
| "epoch": 19.82456140350877, |
| "grad_norm": 0.1891440898180008, |
| "learning_rate": 1.4179448813278484e-08, |
| "loss": 0.0772, |
| "num_input_tokens_seen": 704504, |
| "step": 1130 |
| }, |
| { |
| "epoch": 19.912280701754387, |
| "grad_norm": 0.390571653842926, |
| "learning_rate": 4.218959166932268e-09, |
| "loss": 0.1291, |
| "num_input_tokens_seen": 707992, |
| "step": 1135 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 0.058101434260606766, |
| "learning_rate": 1.1719651499819683e-10, |
| "loss": 0.1104, |
| "num_input_tokens_seen": 711112, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.13817264139652252, |
| "eval_runtime": 0.6612, |
| "eval_samples_per_second": 37.812, |
| "eval_steps_per_second": 10.587, |
| "num_input_tokens_seen": 711112, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "num_input_tokens_seen": 711112, |
| "step": 1140, |
| "total_flos": 3.202189082198016e+16, |
| "train_loss": 0.22204159968124146, |
| "train_runtime": 276.4415, |
| "train_samples_per_second": 16.278, |
| "train_steps_per_second": 4.124 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1140, |
| "num_input_tokens_seen": 711112, |
| "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": 3.202189082198016e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
