| { |
| "best_global_step": 798, |
| "best_metric": 0.13781596720218658, |
| "best_model_checkpoint": "saves_multiple/ia3/llama-3-8b-instruct/train_cb_101112_1760637985/checkpoint-798", |
| "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": 3.035602331161499, |
| "learning_rate": 1.7543859649122807e-06, |
| "loss": 0.9476, |
| "num_input_tokens_seen": 3648, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 3.0266945362091064, |
| "learning_rate": 3.9473684210526315e-06, |
| "loss": 1.1572, |
| "num_input_tokens_seen": 6848, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 2.909323215484619, |
| "learning_rate": 6.140350877192982e-06, |
| "loss": 0.9315, |
| "num_input_tokens_seen": 10592, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 2.9558660984039307, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 1.1626, |
| "num_input_tokens_seen": 13856, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.43859649122807015, |
| "grad_norm": 3.1446194648742676, |
| "learning_rate": 1.0526315789473684e-05, |
| "loss": 0.9928, |
| "num_input_tokens_seen": 17760, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 2.9421627521514893, |
| "learning_rate": 1.2719298245614037e-05, |
| "loss": 1.0745, |
| "num_input_tokens_seen": 20704, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6140350877192983, |
| "grad_norm": 3.227771520614624, |
| "learning_rate": 1.4912280701754386e-05, |
| "loss": 1.2294, |
| "num_input_tokens_seen": 23392, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 2.4900922775268555, |
| "learning_rate": 1.7105263157894737e-05, |
| "loss": 1.0938, |
| "num_input_tokens_seen": 26176, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 3.1620306968688965, |
| "learning_rate": 1.929824561403509e-05, |
| "loss": 1.1273, |
| "num_input_tokens_seen": 29536, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 3.019747495651245, |
| "learning_rate": 2.149122807017544e-05, |
| "loss": 1.3321, |
| "num_input_tokens_seen": 32704, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9649122807017544, |
| "grad_norm": 3.5131232738494873, |
| "learning_rate": 2.368421052631579e-05, |
| "loss": 1.2004, |
| "num_input_tokens_seen": 35392, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.1029865741729736, |
| "eval_runtime": 0.5711, |
| "eval_samples_per_second": 43.778, |
| "eval_steps_per_second": 12.258, |
| "num_input_tokens_seen": 36112, |
| "step": 57 |
| }, |
| { |
| "epoch": 1.0526315789473684, |
| "grad_norm": 3.030216693878174, |
| "learning_rate": 2.5877192982456143e-05, |
| "loss": 1.2236, |
| "num_input_tokens_seen": 38192, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1403508771929824, |
| "grad_norm": 3.1718602180480957, |
| "learning_rate": 2.8070175438596492e-05, |
| "loss": 1.1572, |
| "num_input_tokens_seen": 41424, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.2280701754385965, |
| "grad_norm": 2.3236804008483887, |
| "learning_rate": 3.0263157894736844e-05, |
| "loss": 1.0088, |
| "num_input_tokens_seen": 44496, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.3157894736842106, |
| "grad_norm": 2.9596214294433594, |
| "learning_rate": 3.24561403508772e-05, |
| "loss": 1.198, |
| "num_input_tokens_seen": 47408, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.4035087719298245, |
| "grad_norm": 3.237265110015869, |
| "learning_rate": 3.4649122807017546e-05, |
| "loss": 1.228, |
| "num_input_tokens_seen": 50256, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.4912280701754386, |
| "grad_norm": 2.5169811248779297, |
| "learning_rate": 3.6842105263157895e-05, |
| "loss": 0.9661, |
| "num_input_tokens_seen": 52816, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.5789473684210527, |
| "grad_norm": 2.656912326812744, |
| "learning_rate": 3.9035087719298244e-05, |
| "loss": 1.1447, |
| "num_input_tokens_seen": 55856, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 3.0823001861572266, |
| "learning_rate": 4.12280701754386e-05, |
| "loss": 1.1003, |
| "num_input_tokens_seen": 59920, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.7543859649122808, |
| "grad_norm": 3.5630784034729004, |
| "learning_rate": 4.342105263157895e-05, |
| "loss": 1.1226, |
| "num_input_tokens_seen": 62512, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.8421052631578947, |
| "grad_norm": 3.707685708999634, |
| "learning_rate": 4.56140350877193e-05, |
| "loss": 0.9227, |
| "num_input_tokens_seen": 66288, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.9298245614035088, |
| "grad_norm": 1.9192754030227661, |
| "learning_rate": 4.780701754385965e-05, |
| "loss": 0.657, |
| "num_input_tokens_seen": 69328, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.5552152395248413, |
| "eval_runtime": 0.5759, |
| "eval_samples_per_second": 43.41, |
| "eval_steps_per_second": 12.155, |
| "num_input_tokens_seen": 71552, |
| "step": 114 |
| }, |
| { |
| "epoch": 2.017543859649123, |
| "grad_norm": 2.8221514225006104, |
| "learning_rate": 5e-05, |
| "loss": 0.5586, |
| "num_input_tokens_seen": 71968, |
| "step": 115 |
| }, |
| { |
| "epoch": 2.1052631578947367, |
| "grad_norm": 1.805423617362976, |
| "learning_rate": 4.999707014206475e-05, |
| "loss": 0.3762, |
| "num_input_tokens_seen": 75040, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.192982456140351, |
| "grad_norm": 1.7813746929168701, |
| "learning_rate": 4.9988281254984414e-05, |
| "loss": 0.4646, |
| "num_input_tokens_seen": 78624, |
| "step": 125 |
| }, |
| { |
| "epoch": 2.280701754385965, |
| "grad_norm": 0.6682248115539551, |
| "learning_rate": 4.997363539877422e-05, |
| "loss": 0.1992, |
| "num_input_tokens_seen": 81824, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.3684210526315788, |
| "grad_norm": 0.8201332688331604, |
| "learning_rate": 4.9953136006256415e-05, |
| "loss": 0.1369, |
| "num_input_tokens_seen": 84960, |
| "step": 135 |
| }, |
| { |
| "epoch": 2.456140350877193, |
| "grad_norm": 1.1988774538040161, |
| "learning_rate": 4.9926787882255636e-05, |
| "loss": 0.2092, |
| "num_input_tokens_seen": 89024, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.543859649122807, |
| "grad_norm": 1.2600805759429932, |
| "learning_rate": 4.9894597202472696e-05, |
| "loss": 0.1343, |
| "num_input_tokens_seen": 92480, |
| "step": 145 |
| }, |
| { |
| "epoch": 2.6315789473684212, |
| "grad_norm": 0.21511641144752502, |
| "learning_rate": 4.985657151203706e-05, |
| "loss": 0.111, |
| "num_input_tokens_seen": 95520, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.719298245614035, |
| "grad_norm": 1.118030309677124, |
| "learning_rate": 4.9812719723738435e-05, |
| "loss": 0.2328, |
| "num_input_tokens_seen": 98976, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.807017543859649, |
| "grad_norm": 0.5893439650535583, |
| "learning_rate": 4.976305211593758e-05, |
| "loss": 0.177, |
| "num_input_tokens_seen": 101600, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.8947368421052633, |
| "grad_norm": 0.7333016395568848, |
| "learning_rate": 4.970758033015731e-05, |
| "loss": 0.1671, |
| "num_input_tokens_seen": 104544, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.982456140350877, |
| "grad_norm": 2.9451327323913574, |
| "learning_rate": 4.9646317368353743e-05, |
| "loss": 0.3455, |
| "num_input_tokens_seen": 107936, |
| "step": 170 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.1738632768392563, |
| "eval_runtime": 0.5779, |
| "eval_samples_per_second": 43.263, |
| "eval_steps_per_second": 12.114, |
| "num_input_tokens_seen": 108088, |
| "step": 171 |
| }, |
| { |
| "epoch": 3.0701754385964914, |
| "grad_norm": 1.805781364440918, |
| "learning_rate": 4.957927758986888e-05, |
| "loss": 0.1656, |
| "num_input_tokens_seen": 110360, |
| "step": 175 |
| }, |
| { |
| "epoch": 3.1578947368421053, |
| "grad_norm": 0.2506971061229706, |
| "learning_rate": 4.9506476708064865e-05, |
| "loss": 0.1268, |
| "num_input_tokens_seen": 113656, |
| "step": 180 |
| }, |
| { |
| "epoch": 3.245614035087719, |
| "grad_norm": 0.7638034224510193, |
| "learning_rate": 4.9427931786641e-05, |
| "loss": 0.053, |
| "num_input_tokens_seen": 116792, |
| "step": 185 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 1.6796597242355347, |
| "learning_rate": 4.93436612356342e-05, |
| "loss": 0.1801, |
| "num_input_tokens_seen": 119864, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.4210526315789473, |
| "grad_norm": 0.608403742313385, |
| "learning_rate": 4.925368480710385e-05, |
| "loss": 0.1717, |
| "num_input_tokens_seen": 123512, |
| "step": 195 |
| }, |
| { |
| "epoch": 3.5087719298245617, |
| "grad_norm": 1.9806897640228271, |
| "learning_rate": 4.915802359050222e-05, |
| "loss": 0.2954, |
| "num_input_tokens_seen": 127064, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.5964912280701755, |
| "grad_norm": 0.6469482779502869, |
| "learning_rate": 4.905670000773126e-05, |
| "loss": 0.1505, |
| "num_input_tokens_seen": 130008, |
| "step": 205 |
| }, |
| { |
| "epoch": 3.6842105263157894, |
| "grad_norm": 0.4952060282230377, |
| "learning_rate": 4.894973780788722e-05, |
| "loss": 0.1001, |
| "num_input_tokens_seen": 132760, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.7719298245614032, |
| "grad_norm": 0.33997464179992676, |
| "learning_rate": 4.88371620616941e-05, |
| "loss": 0.0926, |
| "num_input_tokens_seen": 136728, |
| "step": 215 |
| }, |
| { |
| "epoch": 3.8596491228070176, |
| "grad_norm": 0.46155115962028503, |
| "learning_rate": 4.871899915562736e-05, |
| "loss": 0.2277, |
| "num_input_tokens_seen": 139800, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.9473684210526314, |
| "grad_norm": 1.5655906200408936, |
| "learning_rate": 4.8595276785729236e-05, |
| "loss": 0.2157, |
| "num_input_tokens_seen": 143128, |
| "step": 225 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.1595449298620224, |
| "eval_runtime": 0.5743, |
| "eval_samples_per_second": 43.531, |
| "eval_steps_per_second": 12.189, |
| "num_input_tokens_seen": 144720, |
| "step": 228 |
| }, |
| { |
| "epoch": 4.035087719298246, |
| "grad_norm": 0.547755241394043, |
| "learning_rate": 4.846602395111711e-05, |
| "loss": 0.088, |
| "num_input_tokens_seen": 145904, |
| "step": 230 |
| }, |
| { |
| "epoch": 4.12280701754386, |
| "grad_norm": 2.529719591140747, |
| "learning_rate": 4.833127094718643e-05, |
| "loss": 0.2959, |
| "num_input_tokens_seen": 149520, |
| "step": 235 |
| }, |
| { |
| "epoch": 4.2105263157894735, |
| "grad_norm": 0.8846591114997864, |
| "learning_rate": 4.819104935850983e-05, |
| "loss": 0.1018, |
| "num_input_tokens_seen": 152912, |
| "step": 240 |
| }, |
| { |
| "epoch": 4.298245614035087, |
| "grad_norm": 0.12779730558395386, |
| "learning_rate": 4.804539205143405e-05, |
| "loss": 0.0865, |
| "num_input_tokens_seen": 156176, |
| "step": 245 |
| }, |
| { |
| "epoch": 4.385964912280702, |
| "grad_norm": 1.2964743375778198, |
| "learning_rate": 4.789433316637644e-05, |
| "loss": 0.1465, |
| "num_input_tokens_seen": 159376, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.473684210526316, |
| "grad_norm": 2.937610626220703, |
| "learning_rate": 4.7737908109822854e-05, |
| "loss": 0.2832, |
| "num_input_tokens_seen": 162128, |
| "step": 255 |
| }, |
| { |
| "epoch": 4.56140350877193, |
| "grad_norm": 0.15169142186641693, |
| "learning_rate": 4.757615354602874e-05, |
| "loss": 0.0999, |
| "num_input_tokens_seen": 165456, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.649122807017544, |
| "grad_norm": 0.267172247171402, |
| "learning_rate": 4.7409107388425504e-05, |
| "loss": 0.1391, |
| "num_input_tokens_seen": 168272, |
| "step": 265 |
| }, |
| { |
| "epoch": 4.7368421052631575, |
| "grad_norm": 1.3118361234664917, |
| "learning_rate": 4.723680879073396e-05, |
| "loss": 0.1241, |
| "num_input_tokens_seen": 171216, |
| "step": 270 |
| }, |
| { |
| "epoch": 4.824561403508772, |
| "grad_norm": 0.32808059453964233, |
| "learning_rate": 4.70592981377872e-05, |
| "loss": 0.1205, |
| "num_input_tokens_seen": 174192, |
| "step": 275 |
| }, |
| { |
| "epoch": 4.912280701754386, |
| "grad_norm": 0.5601760745048523, |
| "learning_rate": 4.6876617036064844e-05, |
| "loss": 0.1558, |
| "num_input_tokens_seen": 178320, |
| "step": 280 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.53883296251297, |
| "learning_rate": 4.668880830394093e-05, |
| "loss": 0.151, |
| "num_input_tokens_seen": 181120, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.16091668605804443, |
| "eval_runtime": 0.5744, |
| "eval_samples_per_second": 43.527, |
| "eval_steps_per_second": 12.188, |
| "num_input_tokens_seen": 181120, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.087719298245614, |
| "grad_norm": 1.0146218538284302, |
| "learning_rate": 4.649591596164778e-05, |
| "loss": 0.0918, |
| "num_input_tokens_seen": 184224, |
| "step": 290 |
| }, |
| { |
| "epoch": 5.175438596491228, |
| "grad_norm": 0.47441354393959045, |
| "learning_rate": 4.629798522095818e-05, |
| "loss": 0.1552, |
| "num_input_tokens_seen": 187296, |
| "step": 295 |
| }, |
| { |
| "epoch": 5.2631578947368425, |
| "grad_norm": 0.7975168228149414, |
| "learning_rate": 4.6095062474588225e-05, |
| "loss": 0.2883, |
| "num_input_tokens_seen": 190944, |
| "step": 300 |
| }, |
| { |
| "epoch": 5.350877192982456, |
| "grad_norm": 0.6261847019195557, |
| "learning_rate": 4.588719528532342e-05, |
| "loss": 0.1341, |
| "num_input_tokens_seen": 194080, |
| "step": 305 |
| }, |
| { |
| "epoch": 5.43859649122807, |
| "grad_norm": 0.19188463687896729, |
| "learning_rate": 4.5674432374870455e-05, |
| "loss": 0.2912, |
| "num_input_tokens_seen": 197312, |
| "step": 310 |
| }, |
| { |
| "epoch": 5.526315789473684, |
| "grad_norm": 0.2997607886791229, |
| "learning_rate": 4.545682361243748e-05, |
| "loss": 0.0382, |
| "num_input_tokens_seen": 201280, |
| "step": 315 |
| }, |
| { |
| "epoch": 5.614035087719298, |
| "grad_norm": 1.1383190155029297, |
| "learning_rate": 4.5234420003045236e-05, |
| "loss": 0.156, |
| "num_input_tokens_seen": 204256, |
| "step": 320 |
| }, |
| { |
| "epoch": 5.701754385964913, |
| "grad_norm": 0.7752416133880615, |
| "learning_rate": 4.5007273675572104e-05, |
| "loss": 0.0881, |
| "num_input_tokens_seen": 207488, |
| "step": 325 |
| }, |
| { |
| "epoch": 5.7894736842105265, |
| "grad_norm": 0.5944119691848755, |
| "learning_rate": 4.4775437870535685e-05, |
| "loss": 0.1128, |
| "num_input_tokens_seen": 210240, |
| "step": 330 |
| }, |
| { |
| "epoch": 5.87719298245614, |
| "grad_norm": 0.9908325672149658, |
| "learning_rate": 4.4538966927613836e-05, |
| "loss": 0.0951, |
| "num_input_tokens_seen": 213952, |
| "step": 335 |
| }, |
| { |
| "epoch": 5.964912280701754, |
| "grad_norm": 0.12694884836673737, |
| "learning_rate": 4.4297916272908024e-05, |
| "loss": 0.1168, |
| "num_input_tokens_seen": 216352, |
| "step": 340 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.15841087698936462, |
| "eval_runtime": 0.575, |
| "eval_samples_per_second": 43.479, |
| "eval_steps_per_second": 12.174, |
| "num_input_tokens_seen": 217128, |
| "step": 342 |
| }, |
| { |
| "epoch": 6.052631578947368, |
| "grad_norm": 1.1227468252182007, |
| "learning_rate": 4.405234240595214e-05, |
| "loss": 0.1444, |
| "num_input_tokens_seen": 219528, |
| "step": 345 |
| }, |
| { |
| "epoch": 6.140350877192983, |
| "grad_norm": 0.332541286945343, |
| "learning_rate": 4.3802302886469606e-05, |
| "loss": 0.064, |
| "num_input_tokens_seen": 222472, |
| "step": 350 |
| }, |
| { |
| "epoch": 6.228070175438597, |
| "grad_norm": 1.9734841585159302, |
| "learning_rate": 4.3547856320882044e-05, |
| "loss": 0.293, |
| "num_input_tokens_seen": 225448, |
| "step": 355 |
| }, |
| { |
| "epoch": 6.315789473684211, |
| "grad_norm": 1.0871795415878296, |
| "learning_rate": 4.328906234857259e-05, |
| "loss": 0.2759, |
| "num_input_tokens_seen": 228584, |
| "step": 360 |
| }, |
| { |
| "epoch": 6.4035087719298245, |
| "grad_norm": 1.4011313915252686, |
| "learning_rate": 4.302598162790712e-05, |
| "loss": 0.085, |
| "num_input_tokens_seen": 231624, |
| "step": 365 |
| }, |
| { |
| "epoch": 6.491228070175438, |
| "grad_norm": 0.15201807022094727, |
| "learning_rate": 4.27586758220166e-05, |
| "loss": 0.0858, |
| "num_input_tokens_seen": 234856, |
| "step": 370 |
| }, |
| { |
| "epoch": 6.578947368421053, |
| "grad_norm": 1.6477054357528687, |
| "learning_rate": 4.2487207584343955e-05, |
| "loss": 0.1467, |
| "num_input_tokens_seen": 238024, |
| "step": 375 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.5616220235824585, |
| "learning_rate": 4.2211640543958796e-05, |
| "loss": 0.0636, |
| "num_input_tokens_seen": 241576, |
| "step": 380 |
| }, |
| { |
| "epoch": 6.754385964912281, |
| "grad_norm": 0.25579485297203064, |
| "learning_rate": 4.193203929064353e-05, |
| "loss": 0.0393, |
| "num_input_tokens_seen": 244680, |
| "step": 385 |
| }, |
| { |
| "epoch": 6.842105263157895, |
| "grad_norm": 3.6579537391662598, |
| "learning_rate": 4.164846935975421e-05, |
| "loss": 0.2693, |
| "num_input_tokens_seen": 247880, |
| "step": 390 |
| }, |
| { |
| "epoch": 6.9298245614035086, |
| "grad_norm": 1.5050499439239502, |
| "learning_rate": 4.136099721685983e-05, |
| "loss": 0.0967, |
| "num_input_tokens_seen": 251496, |
| "step": 395 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.15077444911003113, |
| "eval_runtime": 0.5758, |
| "eval_samples_per_second": 43.42, |
| "eval_steps_per_second": 12.158, |
| "num_input_tokens_seen": 253536, |
| "step": 399 |
| }, |
| { |
| "epoch": 7.017543859649122, |
| "grad_norm": 0.17453746497631073, |
| "learning_rate": 4.1069690242163484e-05, |
| "loss": 0.0835, |
| "num_input_tokens_seen": 254016, |
| "step": 400 |
| }, |
| { |
| "epoch": 7.105263157894737, |
| "grad_norm": 1.2112236022949219, |
| "learning_rate": 4.0774616714709316e-05, |
| "loss": 0.1209, |
| "num_input_tokens_seen": 257024, |
| "step": 405 |
| }, |
| { |
| "epoch": 7.192982456140351, |
| "grad_norm": 0.25198495388031006, |
| "learning_rate": 4.047584579637857e-05, |
| "loss": 0.0999, |
| "num_input_tokens_seen": 260096, |
| "step": 410 |
| }, |
| { |
| "epoch": 7.280701754385965, |
| "grad_norm": 0.21481497585773468, |
| "learning_rate": 4.0173447515678916e-05, |
| "loss": 0.0691, |
| "num_input_tokens_seen": 263776, |
| "step": 415 |
| }, |
| { |
| "epoch": 7.368421052631579, |
| "grad_norm": 0.5673413872718811, |
| "learning_rate": 3.986749275133057e-05, |
| "loss": 0.1655, |
| "num_input_tokens_seen": 267008, |
| "step": 420 |
| }, |
| { |
| "epoch": 7.456140350877193, |
| "grad_norm": 1.493493676185608, |
| "learning_rate": 3.955805321565304e-05, |
| "loss": 0.2283, |
| "num_input_tokens_seen": 270368, |
| "step": 425 |
| }, |
| { |
| "epoch": 7.543859649122807, |
| "grad_norm": 1.0681108236312866, |
| "learning_rate": 3.9245201437756654e-05, |
| "loss": 0.0939, |
| "num_input_tokens_seen": 273568, |
| "step": 430 |
| }, |
| { |
| "epoch": 7.631578947368421, |
| "grad_norm": 0.8637136220932007, |
| "learning_rate": 3.892901074654255e-05, |
| "loss": 0.0496, |
| "num_input_tokens_seen": 277152, |
| "step": 435 |
| }, |
| { |
| "epoch": 7.719298245614035, |
| "grad_norm": 0.8277631402015686, |
| "learning_rate": 3.860955525351516e-05, |
| "loss": 0.1106, |
| "num_input_tokens_seen": 280640, |
| "step": 440 |
| }, |
| { |
| "epoch": 7.807017543859649, |
| "grad_norm": 1.3838900327682495, |
| "learning_rate": 3.82869098354114e-05, |
| "loss": 0.1969, |
| "num_input_tokens_seen": 283392, |
| "step": 445 |
| }, |
| { |
| "epoch": 7.894736842105263, |
| "grad_norm": 0.2915225625038147, |
| "learning_rate": 3.796115011665034e-05, |
| "loss": 0.2669, |
| "num_input_tokens_seen": 286496, |
| "step": 450 |
| }, |
| { |
| "epoch": 7.982456140350877, |
| "grad_norm": 0.3368154466152191, |
| "learning_rate": 3.763235245160775e-05, |
| "loss": 0.0816, |
| "num_input_tokens_seen": 289952, |
| "step": 455 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.14625966548919678, |
| "eval_runtime": 0.5741, |
| "eval_samples_per_second": 43.546, |
| "eval_steps_per_second": 12.193, |
| "num_input_tokens_seen": 290112, |
| "step": 456 |
| }, |
| { |
| "epoch": 8.070175438596491, |
| "grad_norm": 0.7272406816482544, |
| "learning_rate": 3.7300593906719464e-05, |
| "loss": 0.1353, |
| "num_input_tokens_seen": 292512, |
| "step": 460 |
| }, |
| { |
| "epoch": 8.157894736842104, |
| "grad_norm": 0.700942873954773, |
| "learning_rate": 3.69659522424179e-05, |
| "loss": 0.2953, |
| "num_input_tokens_seen": 295456, |
| "step": 465 |
| }, |
| { |
| "epoch": 8.24561403508772, |
| "grad_norm": 0.1716238409280777, |
| "learning_rate": 3.662850589490592e-05, |
| "loss": 0.0646, |
| "num_input_tokens_seen": 299424, |
| "step": 470 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.5740395188331604, |
| "learning_rate": 3.628833395777224e-05, |
| "loss": 0.1339, |
| "num_input_tokens_seen": 302656, |
| "step": 475 |
| }, |
| { |
| "epoch": 8.421052631578947, |
| "grad_norm": 0.9650786519050598, |
| "learning_rate": 3.59455161634528e-05, |
| "loss": 0.1233, |
| "num_input_tokens_seen": 305952, |
| "step": 480 |
| }, |
| { |
| "epoch": 8.508771929824562, |
| "grad_norm": 0.29233530163764954, |
| "learning_rate": 3.560013286454242e-05, |
| "loss": 0.0582, |
| "num_input_tokens_seen": 309248, |
| "step": 485 |
| }, |
| { |
| "epoch": 8.596491228070175, |
| "grad_norm": 0.5102087259292603, |
| "learning_rate": 3.5252265014961006e-05, |
| "loss": 0.1259, |
| "num_input_tokens_seen": 312224, |
| "step": 490 |
| }, |
| { |
| "epoch": 8.68421052631579, |
| "grad_norm": 0.17764154076576233, |
| "learning_rate": 3.490199415097892e-05, |
| "loss": 0.0981, |
| "num_input_tokens_seen": 315200, |
| "step": 495 |
| }, |
| { |
| "epoch": 8.771929824561404, |
| "grad_norm": 0.4505576193332672, |
| "learning_rate": 3.45494023721058e-05, |
| "loss": 0.1059, |
| "num_input_tokens_seen": 318240, |
| "step": 500 |
| }, |
| { |
| "epoch": 8.859649122807017, |
| "grad_norm": 0.8401296734809875, |
| "learning_rate": 3.4194572321847336e-05, |
| "loss": 0.1973, |
| "num_input_tokens_seen": 321408, |
| "step": 505 |
| }, |
| { |
| "epoch": 8.947368421052632, |
| "grad_norm": 0.06337150186300278, |
| "learning_rate": 3.383758716833459e-05, |
| "loss": 0.0819, |
| "num_input_tokens_seen": 324800, |
| "step": 510 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.14808055758476257, |
| "eval_runtime": 0.5784, |
| "eval_samples_per_second": 43.225, |
| "eval_steps_per_second": 12.103, |
| "num_input_tokens_seen": 325872, |
| "step": 513 |
| }, |
| { |
| "epoch": 9.035087719298245, |
| "grad_norm": 1.3674508333206177, |
| "learning_rate": 3.347853058483037e-05, |
| "loss": 0.0739, |
| "num_input_tokens_seen": 327184, |
| "step": 515 |
| }, |
| { |
| "epoch": 9.12280701754386, |
| "grad_norm": 0.28853166103363037, |
| "learning_rate": 3.311748673011709e-05, |
| "loss": 0.0865, |
| "num_input_tokens_seen": 329744, |
| "step": 520 |
| }, |
| { |
| "epoch": 9.210526315789474, |
| "grad_norm": 1.0980778932571411, |
| "learning_rate": 3.275454022877097e-05, |
| "loss": 0.2092, |
| "num_input_tokens_seen": 332976, |
| "step": 525 |
| }, |
| { |
| "epoch": 9.298245614035087, |
| "grad_norm": 1.3283612728118896, |
| "learning_rate": 3.238977615132697e-05, |
| "loss": 0.1394, |
| "num_input_tokens_seen": 335760, |
| "step": 530 |
| }, |
| { |
| "epoch": 9.385964912280702, |
| "grad_norm": 0.5006604194641113, |
| "learning_rate": 3.202327999433924e-05, |
| "loss": 0.1807, |
| "num_input_tokens_seen": 338128, |
| "step": 535 |
| }, |
| { |
| "epoch": 9.473684210526315, |
| "grad_norm": 0.07026021927595139, |
| "learning_rate": 3.165513766034167e-05, |
| "loss": 0.0734, |
| "num_input_tokens_seen": 342160, |
| "step": 540 |
| }, |
| { |
| "epoch": 9.56140350877193, |
| "grad_norm": 0.09749102592468262, |
| "learning_rate": 3.128543543771336e-05, |
| "loss": 0.0885, |
| "num_input_tokens_seen": 344976, |
| "step": 545 |
| }, |
| { |
| "epoch": 9.649122807017545, |
| "grad_norm": 0.17023645341396332, |
| "learning_rate": 3.091425998045356e-05, |
| "loss": 0.0865, |
| "num_input_tokens_seen": 348208, |
| "step": 550 |
| }, |
| { |
| "epoch": 9.736842105263158, |
| "grad_norm": 1.3468619585037231, |
| "learning_rate": 3.0541698287870965e-05, |
| "loss": 0.1573, |
| "num_input_tokens_seen": 351472, |
| "step": 555 |
| }, |
| { |
| "epoch": 9.824561403508772, |
| "grad_norm": 0.9681034684181213, |
| "learning_rate": 3.01678376841921e-05, |
| "loss": 0.1442, |
| "num_input_tokens_seen": 355152, |
| "step": 560 |
| }, |
| { |
| "epoch": 9.912280701754385, |
| "grad_norm": 0.8875375986099243, |
| "learning_rate": 2.9792765798093465e-05, |
| "loss": 0.1078, |
| "num_input_tokens_seen": 359184, |
| "step": 565 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.09026030451059341, |
| "learning_rate": 2.94165705421624e-05, |
| "loss": 0.0791, |
| "num_input_tokens_seen": 361920, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.14556385576725006, |
| "eval_runtime": 0.576, |
| "eval_samples_per_second": 43.402, |
| "eval_steps_per_second": 12.153, |
| "num_input_tokens_seen": 361920, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.087719298245615, |
| "grad_norm": 1.0138522386550903, |
| "learning_rate": 2.9039340092291373e-05, |
| "loss": 0.1443, |
| "num_input_tokens_seen": 365216, |
| "step": 575 |
| }, |
| { |
| "epoch": 10.175438596491228, |
| "grad_norm": 0.9076150059700012, |
| "learning_rate": 2.8661162867010543e-05, |
| "loss": 0.1158, |
| "num_input_tokens_seen": 368512, |
| "step": 580 |
| }, |
| { |
| "epoch": 10.263157894736842, |
| "grad_norm": 0.9776477217674255, |
| "learning_rate": 2.8282127506763456e-05, |
| "loss": 0.1086, |
| "num_input_tokens_seen": 372032, |
| "step": 585 |
| }, |
| { |
| "epoch": 10.350877192982455, |
| "grad_norm": 0.22465470433235168, |
| "learning_rate": 2.7902322853130757e-05, |
| "loss": 0.1967, |
| "num_input_tokens_seen": 375136, |
| "step": 590 |
| }, |
| { |
| "epoch": 10.43859649122807, |
| "grad_norm": 0.32824060320854187, |
| "learning_rate": 2.752183792800671e-05, |
| "loss": 0.0842, |
| "num_input_tokens_seen": 378208, |
| "step": 595 |
| }, |
| { |
| "epoch": 10.526315789473685, |
| "grad_norm": 0.40594610571861267, |
| "learning_rate": 2.7140761912733474e-05, |
| "loss": 0.1324, |
| "num_input_tokens_seen": 380864, |
| "step": 600 |
| }, |
| { |
| "epoch": 10.614035087719298, |
| "grad_norm": 0.6923454999923706, |
| "learning_rate": 2.6759184127198046e-05, |
| "loss": 0.0615, |
| "num_input_tokens_seen": 384160, |
| "step": 605 |
| }, |
| { |
| "epoch": 10.701754385964913, |
| "grad_norm": 0.28134238719940186, |
| "learning_rate": 2.6377194008896637e-05, |
| "loss": 0.0619, |
| "num_input_tokens_seen": 387584, |
| "step": 610 |
| }, |
| { |
| "epoch": 10.789473684210526, |
| "grad_norm": 0.15947100520133972, |
| "learning_rate": 2.5994881091971605e-05, |
| "loss": 0.1677, |
| "num_input_tokens_seen": 391584, |
| "step": 615 |
| }, |
| { |
| "epoch": 10.87719298245614, |
| "grad_norm": 0.34628206491470337, |
| "learning_rate": 2.5612334986225623e-05, |
| "loss": 0.067, |
| "num_input_tokens_seen": 394592, |
| "step": 620 |
| }, |
| { |
| "epoch": 10.964912280701755, |
| "grad_norm": 0.8037942051887512, |
| "learning_rate": 2.5229645356118163e-05, |
| "loss": 0.1327, |
| "num_input_tokens_seen": 397696, |
| "step": 625 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 0.14215007424354553, |
| "eval_runtime": 0.5753, |
| "eval_samples_per_second": 43.458, |
| "eval_steps_per_second": 12.168, |
| "num_input_tokens_seen": 398432, |
| "step": 627 |
| }, |
| { |
| "epoch": 11.052631578947368, |
| "grad_norm": 0.8208631277084351, |
| "learning_rate": 2.4846901899749185e-05, |
| "loss": 0.1197, |
| "num_input_tokens_seen": 400736, |
| "step": 630 |
| }, |
| { |
| "epoch": 11.140350877192983, |
| "grad_norm": 0.46960461139678955, |
| "learning_rate": 2.4464194327834926e-05, |
| "loss": 0.1065, |
| "num_input_tokens_seen": 403680, |
| "step": 635 |
| }, |
| { |
| "epoch": 11.228070175438596, |
| "grad_norm": 0.6902840733528137, |
| "learning_rate": 2.4081612342680694e-05, |
| "loss": 0.0788, |
| "num_input_tokens_seen": 406784, |
| "step": 640 |
| }, |
| { |
| "epoch": 11.31578947368421, |
| "grad_norm": 0.11541532725095749, |
| "learning_rate": 2.369924561715569e-05, |
| "loss": 0.0949, |
| "num_input_tokens_seen": 410208, |
| "step": 645 |
| }, |
| { |
| "epoch": 11.403508771929825, |
| "grad_norm": 0.468565434217453, |
| "learning_rate": 2.3317183773674718e-05, |
| "loss": 0.1667, |
| "num_input_tokens_seen": 414016, |
| "step": 650 |
| }, |
| { |
| "epoch": 11.491228070175438, |
| "grad_norm": 0.5162321925163269, |
| "learning_rate": 2.2935516363191693e-05, |
| "loss": 0.0709, |
| "num_input_tokens_seen": 416992, |
| "step": 655 |
| }, |
| { |
| "epoch": 11.578947368421053, |
| "grad_norm": 0.1968168318271637, |
| "learning_rate": 2.2554332844209904e-05, |
| "loss": 0.0432, |
| "num_input_tokens_seen": 420224, |
| "step": 660 |
| }, |
| { |
| "epoch": 11.666666666666666, |
| "grad_norm": 1.2083020210266113, |
| "learning_rate": 2.2173722561813987e-05, |
| "loss": 0.0993, |
| "num_input_tokens_seen": 423456, |
| "step": 665 |
| }, |
| { |
| "epoch": 11.75438596491228, |
| "grad_norm": 0.10667666047811508, |
| "learning_rate": 2.179377472672842e-05, |
| "loss": 0.4389, |
| "num_input_tokens_seen": 426560, |
| "step": 670 |
| }, |
| { |
| "epoch": 11.842105263157894, |
| "grad_norm": 0.47717732191085815, |
| "learning_rate": 2.1414578394407597e-05, |
| "loss": 0.0965, |
| "num_input_tokens_seen": 429920, |
| "step": 675 |
| }, |
| { |
| "epoch": 11.929824561403509, |
| "grad_norm": 0.42452216148376465, |
| "learning_rate": 2.1036222444162147e-05, |
| "loss": 0.0426, |
| "num_input_tokens_seen": 433088, |
| "step": 680 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 0.1419922560453415, |
| "eval_runtime": 0.5766, |
| "eval_samples_per_second": 43.358, |
| "eval_steps_per_second": 12.14, |
| "num_input_tokens_seen": 435536, |
| "step": 684 |
| }, |
| { |
| "epoch": 12.017543859649123, |
| "grad_norm": 0.9932581186294556, |
| "learning_rate": 2.0658795558326743e-05, |
| "loss": 0.1357, |
| "num_input_tokens_seen": 436304, |
| "step": 685 |
| }, |
| { |
| "epoch": 12.105263157894736, |
| "grad_norm": 1.4266047477722168, |
| "learning_rate": 2.0282386201473894e-05, |
| "loss": 0.1389, |
| "num_input_tokens_seen": 439472, |
| "step": 690 |
| }, |
| { |
| "epoch": 12.192982456140351, |
| "grad_norm": 0.8649075031280518, |
| "learning_rate": 1.99070825996789e-05, |
| "loss": 0.0298, |
| "num_input_tokens_seen": 442160, |
| "step": 695 |
| }, |
| { |
| "epoch": 12.280701754385966, |
| "grad_norm": 0.49387601017951965, |
| "learning_rate": 1.9532972719840607e-05, |
| "loss": 0.1005, |
| "num_input_tokens_seen": 445488, |
| "step": 700 |
| }, |
| { |
| "epoch": 12.368421052631579, |
| "grad_norm": 1.2329332828521729, |
| "learning_rate": 1.9160144249063035e-05, |
| "loss": 0.088, |
| "num_input_tokens_seen": 448784, |
| "step": 705 |
| }, |
| { |
| "epoch": 12.456140350877194, |
| "grad_norm": 0.0824417695403099, |
| "learning_rate": 1.8788684574102467e-05, |
| "loss": 0.0405, |
| "num_input_tokens_seen": 452656, |
| "step": 710 |
| }, |
| { |
| "epoch": 12.543859649122806, |
| "grad_norm": 0.9820500612258911, |
| "learning_rate": 1.8418680760885027e-05, |
| "loss": 0.1504, |
| "num_input_tokens_seen": 455888, |
| "step": 715 |
| }, |
| { |
| "epoch": 12.631578947368421, |
| "grad_norm": 0.21919409930706024, |
| "learning_rate": 1.805021953409934e-05, |
| "loss": 0.0751, |
| "num_input_tokens_seen": 458896, |
| "step": 720 |
| }, |
| { |
| "epoch": 12.719298245614034, |
| "grad_norm": 1.0950260162353516, |
| "learning_rate": 1.7683387256869353e-05, |
| "loss": 0.0538, |
| "num_input_tokens_seen": 461648, |
| "step": 725 |
| }, |
| { |
| "epoch": 12.807017543859649, |
| "grad_norm": 1.532854437828064, |
| "learning_rate": 1.7318269910511736e-05, |
| "loss": 0.2139, |
| "num_input_tokens_seen": 464912, |
| "step": 730 |
| }, |
| { |
| "epoch": 12.894736842105264, |
| "grad_norm": 0.7314798831939697, |
| "learning_rate": 1.6954953074382863e-05, |
| "loss": 0.2633, |
| "num_input_tokens_seen": 467888, |
| "step": 735 |
| }, |
| { |
| "epoch": 12.982456140350877, |
| "grad_norm": 0.3990841209888458, |
| "learning_rate": 1.659352190581993e-05, |
| "loss": 0.0717, |
| "num_input_tokens_seen": 471440, |
| "step": 740 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_loss": 0.14322011172771454, |
| "eval_runtime": 0.5742, |
| "eval_samples_per_second": 43.536, |
| "eval_steps_per_second": 12.19, |
| "num_input_tokens_seen": 471520, |
| "step": 741 |
| }, |
| { |
| "epoch": 13.070175438596491, |
| "grad_norm": 0.19514040648937225, |
| "learning_rate": 1.6234061120181142e-05, |
| "loss": 0.0848, |
| "num_input_tokens_seen": 473600, |
| "step": 745 |
| }, |
| { |
| "epoch": 13.157894736842104, |
| "grad_norm": 0.6932633519172668, |
| "learning_rate": 1.5876654970989308e-05, |
| "loss": 0.0807, |
| "num_input_tokens_seen": 476576, |
| "step": 750 |
| }, |
| { |
| "epoch": 13.24561403508772, |
| "grad_norm": 0.15419046580791473, |
| "learning_rate": 1.552138723018382e-05, |
| "loss": 0.0402, |
| "num_input_tokens_seen": 479776, |
| "step": 755 |
| }, |
| { |
| "epoch": 13.333333333333334, |
| "grad_norm": 0.8002160787582397, |
| "learning_rate": 1.5168341168485423e-05, |
| "loss": 0.2385, |
| "num_input_tokens_seen": 482784, |
| "step": 760 |
| }, |
| { |
| "epoch": 13.421052631578947, |
| "grad_norm": 0.22885198891162872, |
| "learning_rate": 1.4817599535878565e-05, |
| "loss": 0.0695, |
| "num_input_tokens_seen": 485696, |
| "step": 765 |
| }, |
| { |
| "epoch": 13.508771929824562, |
| "grad_norm": 0.5722417235374451, |
| "learning_rate": 1.4469244542215682e-05, |
| "loss": 0.0389, |
| "num_input_tokens_seen": 488832, |
| "step": 770 |
| }, |
| { |
| "epoch": 13.596491228070175, |
| "grad_norm": 0.11324457079172134, |
| "learning_rate": 1.4123357837948175e-05, |
| "loss": 0.0761, |
| "num_input_tokens_seen": 492896, |
| "step": 775 |
| }, |
| { |
| "epoch": 13.68421052631579, |
| "grad_norm": 1.4171782732009888, |
| "learning_rate": 1.3780020494988446e-05, |
| "loss": 0.1116, |
| "num_input_tokens_seen": 496224, |
| "step": 780 |
| }, |
| { |
| "epoch": 13.771929824561404, |
| "grad_norm": 0.2043789178133011, |
| "learning_rate": 1.3439312987707615e-05, |
| "loss": 0.2338, |
| "num_input_tokens_seen": 499040, |
| "step": 785 |
| }, |
| { |
| "epoch": 13.859649122807017, |
| "grad_norm": 0.21953469514846802, |
| "learning_rate": 1.3101315174073162e-05, |
| "loss": 0.1554, |
| "num_input_tokens_seen": 502560, |
| "step": 790 |
| }, |
| { |
| "epoch": 13.947368421052632, |
| "grad_norm": 0.7600892782211304, |
| "learning_rate": 1.2766106276931223e-05, |
| "loss": 0.1513, |
| "num_input_tokens_seen": 505152, |
| "step": 795 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_loss": 0.13781596720218658, |
| "eval_runtime": 0.5775, |
| "eval_samples_per_second": 43.294, |
| "eval_steps_per_second": 12.122, |
| "num_input_tokens_seen": 507256, |
| "step": 798 |
| }, |
| { |
| "epoch": 14.035087719298245, |
| "grad_norm": 0.1863660365343094, |
| "learning_rate": 1.243376486543755e-05, |
| "loss": 0.0758, |
| "num_input_tokens_seen": 508888, |
| "step": 800 |
| }, |
| { |
| "epoch": 14.12280701754386, |
| "grad_norm": 0.20367327332496643, |
| "learning_rate": 1.2104368836641908e-05, |
| "loss": 0.0612, |
| "num_input_tokens_seen": 511576, |
| "step": 805 |
| }, |
| { |
| "epoch": 14.210526315789474, |
| "grad_norm": 0.3124436140060425, |
| "learning_rate": 1.1777995397229771e-05, |
| "loss": 0.0725, |
| "num_input_tokens_seen": 514968, |
| "step": 810 |
| }, |
| { |
| "epoch": 14.298245614035087, |
| "grad_norm": 0.4617112874984741, |
| "learning_rate": 1.1454721045426073e-05, |
| "loss": 0.1064, |
| "num_input_tokens_seen": 517912, |
| "step": 815 |
| }, |
| { |
| "epoch": 14.385964912280702, |
| "grad_norm": 0.28274524211883545, |
| "learning_rate": 1.113462155306478e-05, |
| "loss": 0.0803, |
| "num_input_tokens_seen": 521144, |
| "step": 820 |
| }, |
| { |
| "epoch": 14.473684210526315, |
| "grad_norm": 0.5946412682533264, |
| "learning_rate": 1.0817771947828934e-05, |
| "loss": 0.0841, |
| "num_input_tokens_seen": 524152, |
| "step": 825 |
| }, |
| { |
| "epoch": 14.56140350877193, |
| "grad_norm": 0.9001612663269043, |
| "learning_rate": 1.0504246495664932e-05, |
| "loss": 0.1091, |
| "num_input_tokens_seen": 526808, |
| "step": 830 |
| }, |
| { |
| "epoch": 14.649122807017545, |
| "grad_norm": 2.992017984390259, |
| "learning_rate": 1.0194118683375503e-05, |
| "loss": 0.2687, |
| "num_input_tokens_seen": 530072, |
| "step": 835 |
| }, |
| { |
| "epoch": 14.736842105263158, |
| "grad_norm": 1.1040074825286865, |
| "learning_rate": 9.887461201395176e-06, |
| "loss": 0.0908, |
| "num_input_tokens_seen": 534104, |
| "step": 840 |
| }, |
| { |
| "epoch": 14.824561403508772, |
| "grad_norm": 0.5093861222267151, |
| "learning_rate": 9.584345926752524e-06, |
| "loss": 0.1053, |
| "num_input_tokens_seen": 537752, |
| "step": 845 |
| }, |
| { |
| "epoch": 14.912280701754385, |
| "grad_norm": 0.5943445563316345, |
| "learning_rate": 9.284843906222948e-06, |
| "loss": 0.0374, |
| "num_input_tokens_seen": 540696, |
| "step": 850 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 0.16984853148460388, |
| "learning_rate": 8.98902533967618e-06, |
| "loss": 0.2993, |
| "num_input_tokens_seen": 543064, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.13971829414367676, |
| "eval_runtime": 0.5762, |
| "eval_samples_per_second": 43.385, |
| "eval_steps_per_second": 12.148, |
| "num_input_tokens_seen": 543064, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.087719298245615, |
| "grad_norm": 0.31223490834236145, |
| "learning_rate": 8.696959563622174e-06, |
| "loss": 0.0648, |
| "num_input_tokens_seen": 546104, |
| "step": 860 |
| }, |
| { |
| "epoch": 15.175438596491228, |
| "grad_norm": 0.9561710953712463, |
| "learning_rate": 8.40871503495947e-06, |
| "loss": 0.1482, |
| "num_input_tokens_seen": 549112, |
| "step": 865 |
| }, |
| { |
| "epoch": 15.263157894736842, |
| "grad_norm": 1.1612529754638672, |
| "learning_rate": 8.124359314929622e-06, |
| "loss": 0.0837, |
| "num_input_tokens_seen": 552696, |
| "step": 870 |
| }, |
| { |
| "epoch": 15.350877192982455, |
| "grad_norm": 1.1817083358764648, |
| "learning_rate": 7.843959053281663e-06, |
| "loss": 0.0888, |
| "num_input_tokens_seen": 556152, |
| "step": 875 |
| }, |
| { |
| "epoch": 15.43859649122807, |
| "grad_norm": 0.5830992460250854, |
| "learning_rate": 7.5675799726501155e-06, |
| "loss": 0.061, |
| "num_input_tokens_seen": 558936, |
| "step": 880 |
| }, |
| { |
| "epoch": 15.526315789473685, |
| "grad_norm": 0.279432475566864, |
| "learning_rate": 7.295286853150391e-06, |
| "loss": 0.1459, |
| "num_input_tokens_seen": 562584, |
| "step": 885 |
| }, |
| { |
| "epoch": 15.614035087719298, |
| "grad_norm": 1.0549548864364624, |
| "learning_rate": 7.027143517195023e-06, |
| "loss": 0.1006, |
| "num_input_tokens_seen": 566456, |
| "step": 890 |
| }, |
| { |
| "epoch": 15.701754385964913, |
| "grad_norm": 0.791029691696167, |
| "learning_rate": 6.763212814534484e-06, |
| "loss": 0.1409, |
| "num_input_tokens_seen": 569176, |
| "step": 895 |
| }, |
| { |
| "epoch": 15.789473684210526, |
| "grad_norm": 0.36763834953308105, |
| "learning_rate": 6.503556607525838e-06, |
| "loss": 0.0272, |
| "num_input_tokens_seen": 572696, |
| "step": 900 |
| }, |
| { |
| "epoch": 15.87719298245614, |
| "grad_norm": 0.19927436113357544, |
| "learning_rate": 6.248235756632984e-06, |
| "loss": 0.1128, |
| "num_input_tokens_seen": 575832, |
| "step": 905 |
| }, |
| { |
| "epoch": 15.964912280701755, |
| "grad_norm": 0.80943763256073, |
| "learning_rate": 5.997310106161589e-06, |
| "loss": 0.2543, |
| "num_input_tokens_seen": 578808, |
| "step": 910 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_loss": 0.13954666256904602, |
| "eval_runtime": 0.5765, |
| "eval_samples_per_second": 43.368, |
| "eval_steps_per_second": 12.143, |
| "num_input_tokens_seen": 579704, |
| "step": 912 |
| }, |
| { |
| "epoch": 16.05263157894737, |
| "grad_norm": 0.37965354323387146, |
| "learning_rate": 5.7508384702323226e-06, |
| "loss": 0.0439, |
| "num_input_tokens_seen": 581720, |
| "step": 915 |
| }, |
| { |
| "epoch": 16.140350877192983, |
| "grad_norm": 2.0723116397857666, |
| "learning_rate": 5.508878618995439e-06, |
| "loss": 0.1695, |
| "num_input_tokens_seen": 585624, |
| "step": 920 |
| }, |
| { |
| "epoch": 16.228070175438596, |
| "grad_norm": 0.47392889857292175, |
| "learning_rate": 5.271487265090163e-06, |
| "loss": 0.0386, |
| "num_input_tokens_seen": 588408, |
| "step": 925 |
| }, |
| { |
| "epoch": 16.31578947368421, |
| "grad_norm": 1.3424001932144165, |
| "learning_rate": 5.038720050351842e-06, |
| "loss": 0.1996, |
| "num_input_tokens_seen": 591512, |
| "step": 930 |
| }, |
| { |
| "epoch": 16.403508771929825, |
| "grad_norm": 0.24867534637451172, |
| "learning_rate": 4.810631532770182e-06, |
| "loss": 0.0348, |
| "num_input_tokens_seen": 595288, |
| "step": 935 |
| }, |
| { |
| "epoch": 16.49122807017544, |
| "grad_norm": 0.43256276845932007, |
| "learning_rate": 4.587275173701428e-06, |
| "loss": 0.0465, |
| "num_input_tokens_seen": 598360, |
| "step": 940 |
| }, |
| { |
| "epoch": 16.57894736842105, |
| "grad_norm": 0.568631112575531, |
| "learning_rate": 4.368703325337667e-06, |
| "loss": 0.292, |
| "num_input_tokens_seen": 601944, |
| "step": 945 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 0.16130073368549347, |
| "learning_rate": 4.154967218436037e-06, |
| "loss": 0.1513, |
| "num_input_tokens_seen": 604824, |
| "step": 950 |
| }, |
| { |
| "epoch": 16.75438596491228, |
| "grad_norm": 1.0300438404083252, |
| "learning_rate": 3.94611695031086e-06, |
| "loss": 0.0454, |
| "num_input_tokens_seen": 607608, |
| "step": 955 |
| }, |
| { |
| "epoch": 16.842105263157894, |
| "grad_norm": 0.44260847568511963, |
| "learning_rate": 3.74220147309135e-06, |
| "loss": 0.0821, |
| "num_input_tokens_seen": 611128, |
| "step": 960 |
| }, |
| { |
| "epoch": 16.92982456140351, |
| "grad_norm": 1.1428203582763672, |
| "learning_rate": 3.543268582247844e-06, |
| "loss": 0.1008, |
| "num_input_tokens_seen": 614040, |
| "step": 965 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_loss": 0.13944916427135468, |
| "eval_runtime": 0.5762, |
| "eval_samples_per_second": 43.39, |
| "eval_steps_per_second": 12.149, |
| "num_input_tokens_seen": 615960, |
| "step": 969 |
| }, |
| { |
| "epoch": 17.017543859649123, |
| "grad_norm": 0.26485544443130493, |
| "learning_rate": 3.3493649053890326e-06, |
| "loss": 0.0453, |
| "num_input_tokens_seen": 616600, |
| "step": 970 |
| }, |
| { |
| "epoch": 17.105263157894736, |
| "grad_norm": 0.7070243954658508, |
| "learning_rate": 3.1605358913330385e-06, |
| "loss": 0.0975, |
| "num_input_tokens_seen": 619736, |
| "step": 975 |
| }, |
| { |
| "epoch": 17.19298245614035, |
| "grad_norm": 0.5280638337135315, |
| "learning_rate": 2.9768257994546662e-06, |
| "loss": 0.1138, |
| "num_input_tokens_seen": 622680, |
| "step": 980 |
| }, |
| { |
| "epoch": 17.280701754385966, |
| "grad_norm": 0.49299854040145874, |
| "learning_rate": 2.7982776893115627e-06, |
| "loss": 0.0431, |
| "num_input_tokens_seen": 626168, |
| "step": 985 |
| }, |
| { |
| "epoch": 17.36842105263158, |
| "grad_norm": 0.7734525799751282, |
| "learning_rate": 2.624933410551508e-06, |
| "loss": 0.1894, |
| "num_input_tokens_seen": 628952, |
| "step": 990 |
| }, |
| { |
| "epoch": 17.45614035087719, |
| "grad_norm": 1.0822232961654663, |
| "learning_rate": 2.456833593103361e-06, |
| "loss": 0.1189, |
| "num_input_tokens_seen": 632184, |
| "step": 995 |
| }, |
| { |
| "epoch": 17.54385964912281, |
| "grad_norm": 0.7628403306007385, |
| "learning_rate": 2.2940176376538445e-06, |
| "loss": 0.199, |
| "num_input_tokens_seen": 635448, |
| "step": 1000 |
| }, |
| { |
| "epoch": 17.63157894736842, |
| "grad_norm": 0.20374268293380737, |
| "learning_rate": 2.136523706412477e-06, |
| "loss": 0.0855, |
| "num_input_tokens_seen": 638616, |
| "step": 1005 |
| }, |
| { |
| "epoch": 17.719298245614034, |
| "grad_norm": 0.09109702706336975, |
| "learning_rate": 1.984388714166799e-06, |
| "loss": 0.1369, |
| "num_input_tokens_seen": 641688, |
| "step": 1010 |
| }, |
| { |
| "epoch": 17.80701754385965, |
| "grad_norm": 0.47799646854400635, |
| "learning_rate": 1.837648319629956e-06, |
| "loss": 0.1551, |
| "num_input_tokens_seen": 644472, |
| "step": 1015 |
| }, |
| { |
| "epoch": 17.894736842105264, |
| "grad_norm": 0.2836926579475403, |
| "learning_rate": 1.6963369170826943e-06, |
| "loss": 0.0774, |
| "num_input_tokens_seen": 648312, |
| "step": 1020 |
| }, |
| { |
| "epoch": 17.982456140350877, |
| "grad_norm": 0.18306328356266022, |
| "learning_rate": 1.5604876283117326e-06, |
| "loss": 0.0189, |
| "num_input_tokens_seen": 652184, |
| "step": 1025 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_loss": 0.14001759886741638, |
| "eval_runtime": 0.5754, |
| "eval_samples_per_second": 43.451, |
| "eval_steps_per_second": 12.166, |
| "num_input_tokens_seen": 652368, |
| "step": 1026 |
| }, |
| { |
| "epoch": 18.07017543859649, |
| "grad_norm": 0.17515921592712402, |
| "learning_rate": 1.4301322948464147e-06, |
| "loss": 0.136, |
| "num_input_tokens_seen": 655312, |
| "step": 1030 |
| }, |
| { |
| "epoch": 18.157894736842106, |
| "grad_norm": 0.3242197632789612, |
| "learning_rate": 1.3053014704953987e-06, |
| "loss": 0.1032, |
| "num_input_tokens_seen": 658896, |
| "step": 1035 |
| }, |
| { |
| "epoch": 18.24561403508772, |
| "grad_norm": 0.2218325436115265, |
| "learning_rate": 1.1860244141851773e-06, |
| "loss": 0.0912, |
| "num_input_tokens_seen": 662288, |
| "step": 1040 |
| }, |
| { |
| "epoch": 18.333333333333332, |
| "grad_norm": 0.1903427094221115, |
| "learning_rate": 1.0723290831021471e-06, |
| "loss": 0.0136, |
| "num_input_tokens_seen": 665104, |
| "step": 1045 |
| }, |
| { |
| "epoch": 18.42105263157895, |
| "grad_norm": 0.337074339389801, |
| "learning_rate": 9.642421261397472e-07, |
| "loss": 0.1666, |
| "num_input_tokens_seen": 667536, |
| "step": 1050 |
| }, |
| { |
| "epoch": 18.50877192982456, |
| "grad_norm": 1.0385386943817139, |
| "learning_rate": 8.617888776522642e-07, |
| "loss": 0.091, |
| "num_input_tokens_seen": 670448, |
| "step": 1055 |
| }, |
| { |
| "epoch": 18.596491228070175, |
| "grad_norm": 1.1613856554031372, |
| "learning_rate": 7.649933515167407e-07, |
| "loss": 0.1429, |
| "num_input_tokens_seen": 674192, |
| "step": 1060 |
| }, |
| { |
| "epoch": 18.68421052631579, |
| "grad_norm": 0.8920570015907288, |
| "learning_rate": 6.738782355044049e-07, |
| "loss": 0.1059, |
| "num_input_tokens_seen": 676784, |
| "step": 1065 |
| }, |
| { |
| "epoch": 18.771929824561404, |
| "grad_norm": 1.2733691930770874, |
| "learning_rate": 5.88464885962911e-07, |
| "loss": 0.2846, |
| "num_input_tokens_seen": 679280, |
| "step": 1070 |
| }, |
| { |
| "epoch": 18.859649122807017, |
| "grad_norm": 0.3717803955078125, |
| "learning_rate": 5.087733228106517e-07, |
| "loss": 0.0494, |
| "num_input_tokens_seen": 682160, |
| "step": 1075 |
| }, |
| { |
| "epoch": 18.94736842105263, |
| "grad_norm": 0.1914367377758026, |
| "learning_rate": 4.3482222484432513e-07, |
| "loss": 0.0258, |
| "num_input_tokens_seen": 685968, |
| "step": 1080 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_loss": 0.13885821402072906, |
| "eval_runtime": 0.5798, |
| "eval_samples_per_second": 43.118, |
| "eval_steps_per_second": 12.073, |
| "num_input_tokens_seen": 687976, |
| "step": 1083 |
| }, |
| { |
| "epoch": 19.035087719298247, |
| "grad_norm": 0.42715832591056824, |
| "learning_rate": 3.666289253608235e-07, |
| "loss": 0.1042, |
| "num_input_tokens_seen": 689128, |
| "step": 1085 |
| }, |
| { |
| "epoch": 19.12280701754386, |
| "grad_norm": 3.0042383670806885, |
| "learning_rate": 3.0420940809451624e-07, |
| "loss": 0.2229, |
| "num_input_tokens_seen": 692136, |
| "step": 1090 |
| }, |
| { |
| "epoch": 19.210526315789473, |
| "grad_norm": 0.689067006111145, |
| "learning_rate": 2.47578303470844e-07, |
| "loss": 0.1107, |
| "num_input_tokens_seen": 695112, |
| "step": 1095 |
| }, |
| { |
| "epoch": 19.29824561403509, |
| "grad_norm": 0.08937988430261612, |
| "learning_rate": 1.96748885177106e-07, |
| "loss": 0.0599, |
| "num_input_tokens_seen": 698600, |
| "step": 1100 |
| }, |
| { |
| "epoch": 19.385964912280702, |
| "grad_norm": 1.0284936428070068, |
| "learning_rate": 1.517330670512629e-07, |
| "loss": 0.2583, |
| "num_input_tokens_seen": 701640, |
| "step": 1105 |
| }, |
| { |
| "epoch": 19.473684210526315, |
| "grad_norm": 0.8484510779380798, |
| "learning_rate": 1.125414002894759e-07, |
| "loss": 0.0761, |
| "num_input_tokens_seen": 704488, |
| "step": 1110 |
| }, |
| { |
| "epoch": 19.56140350877193, |
| "grad_norm": 0.40427008271217346, |
| "learning_rate": 7.918307097301014e-08, |
| "loss": 0.0294, |
| "num_input_tokens_seen": 707848, |
| "step": 1115 |
| }, |
| { |
| "epoch": 19.649122807017545, |
| "grad_norm": 1.0363014936447144, |
| "learning_rate": 5.166589791513465e-08, |
| "loss": 0.0268, |
| "num_input_tokens_seen": 711112, |
| "step": 1120 |
| }, |
| { |
| "epoch": 19.736842105263158, |
| "grad_norm": 0.6243547797203064, |
| "learning_rate": 2.999633082847453e-08, |
| "loss": 0.0742, |
| "num_input_tokens_seen": 714344, |
| "step": 1125 |
| }, |
| { |
| "epoch": 19.82456140350877, |
| "grad_norm": 0.5263058543205261, |
| "learning_rate": 1.4179448813278484e-08, |
| "loss": 0.042, |
| "num_input_tokens_seen": 718280, |
| "step": 1130 |
| }, |
| { |
| "epoch": 19.912280701754387, |
| "grad_norm": 1.756091594696045, |
| "learning_rate": 4.218959166932268e-09, |
| "loss": 0.1362, |
| "num_input_tokens_seen": 720968, |
| "step": 1135 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 0.23448294401168823, |
| "learning_rate": 1.1719651499819683e-10, |
| "loss": 0.1754, |
| "num_input_tokens_seen": 723584, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.1403312385082245, |
| "eval_runtime": 0.5799, |
| "eval_samples_per_second": 43.109, |
| "eval_steps_per_second": 12.071, |
| "num_input_tokens_seen": 723584, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "num_input_tokens_seen": 723584, |
| "step": 1140, |
| "total_flos": 3.258351405760512e+16, |
| "train_loss": 0.22416574166140013, |
| "train_runtime": 273.422, |
| "train_samples_per_second": 16.458, |
| "train_steps_per_second": 4.169 |
| } |
| ], |
| "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": 3.258351405760512e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
