| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 96, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.7632, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.7452, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1e-05, | |
| "loss": 1.707, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.99714745464859e-06, | |
| "loss": 1.5932, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.988593073400354e-06, | |
| "loss": 1.4332, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.974346616959476e-06, | |
| "loss": 1.3474, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.954424340791195e-06, | |
| "loss": 1.348, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.92884897657402e-06, | |
| "loss": 1.3613, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.897649706262474e-06, | |
| "loss": 1.3172, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.860862128789954e-06, | |
| "loss": 1.2214, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 9.818528219449705e-06, | |
| "loss": 1.2694, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 9.770696282000245e-06, | |
| "loss": 1.2959, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 9.717420893549902e-06, | |
| "loss": 1.2342, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.658762842283343e-06, | |
| "loss": 1.2256, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 9.594789058101154e-06, | |
| "loss": 1.1465, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.525572536251608e-06, | |
| "loss": 1.2182, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 9.451192254041759e-06, | |
| "loss": 1.1945, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 9.371733080722911e-06, | |
| "loss": 1.2153, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 9.287285680653254e-06, | |
| "loss": 1.1858, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 9.197946409848196e-06, | |
| "loss": 1.1714, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 9.103817206036383e-06, | |
| "loss": 1.1375, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.005005472346923e-06, | |
| "loss": 1.1346, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 8.90162395476046e-06, | |
| "loss": 1.1966, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.793790613463956e-06, | |
| "loss": 1.1853, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 8.681628488255986e-06, | |
| "loss": 1.1598, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 8.565265558156101e-06, | |
| "loss": 1.0983, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 8.444834595378434e-06, | |
| "loss": 1.0845, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 8.320473013836197e-06, | |
| "loss": 1.1537, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 8.192322712349917e-06, | |
| "loss": 1.1033, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 8.060529912738316e-06, | |
| "loss": 1.1635, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 7.925244992976538e-06, | |
| "loss": 1.1818, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 7.786622315612182e-06, | |
| "loss": 1.0385, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 7.644820051634813e-06, | |
| "loss": 0.9627, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.9101, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 7.352327403013779e-06, | |
| "loss": 0.8977, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 7.201970757788172e-06, | |
| "loss": 0.8862, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 7.049101623982938e-06, | |
| "loss": 0.9568, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 6.893894428052881e-06, | |
| "loss": 0.9041, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 6.736526264224101e-06, | |
| "loss": 0.879, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 6.5771766924262795e-06, | |
| "loss": 0.8772, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 6.41602753341152e-06, | |
| "loss": 0.8311, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 6.2532626612936035e-06, | |
| "loss": 0.835, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 6.089067793744258e-06, | |
| "loss": 0.8575, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 5.923630280085948e-06, | |
| "loss": 0.8358, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.757138887522884e-06, | |
| "loss": 0.9031, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.5897835857542315e-06, | |
| "loss": 0.8421, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.421755330215223e-06, | |
| "loss": 0.8405, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 5.253245844193564e-06, | |
| "loss": 0.8791, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 5.084447400069656e-06, | |
| "loss": 0.822, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.915552599930345e-06, | |
| "loss": 0.8371, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.746754155806437e-06, | |
| "loss": 0.7871, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.5782446697847775e-06, | |
| "loss": 0.8458, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 4.410216414245771e-06, | |
| "loss": 0.853, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.2428611124771184e-06, | |
| "loss": 0.8843, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.076369719914055e-06, | |
| "loss": 0.8413, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 3.910932206255742e-06, | |
| "loss": 0.8236, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 3.7467373387063973e-06, | |
| "loss": 0.8259, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 3.58397246658848e-06, | |
| "loss": 0.8495, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.4228233075737225e-06, | |
| "loss": 0.8405, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 3.2634737357758994e-06, | |
| "loss": 0.8137, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 3.10610557194712e-06, | |
| "loss": 0.8442, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 2.950898376017064e-06, | |
| "loss": 0.8286, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.7980292422118282e-06, | |
| "loss": 0.8154, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.6476725969862227e-06, | |
| "loss": 0.7476, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.6805, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 2.3551799483651894e-06, | |
| "loss": 0.7016, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 2.2133776843878185e-06, | |
| "loss": 0.6793, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.074755007023461e-06, | |
| "loss": 0.6642, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.9394700872616856e-06, | |
| "loss": 0.6314, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.8076772876500831e-06, | |
| "loss": 0.6427, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.6795269861638041e-06, | |
| "loss": 0.6854, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.555165404621567e-06, | |
| "loss": 0.6639, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.434734441843899e-06, | |
| "loss": 0.6829, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.3183715117440143e-06, | |
| "loss": 0.653, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.2062093865360458e-06, | |
| "loss": 0.6338, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.0983760452395415e-06, | |
| "loss": 0.6421, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 9.949945276530782e-07, | |
| "loss": 0.6437, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 8.961827939636198e-07, | |
| "loss": 0.6448, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 8.02053590151805e-07, | |
| "loss": 0.6357, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 7.127143193467445e-07, | |
| "loss": 0.6663, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 6.282669192770896e-07, | |
| "loss": 0.6638, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 5.488077459582425e-07, | |
| "loss": 0.6636, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.7442746374839363e-07, | |
| "loss": 0.6258, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.05210941898847e-07, | |
| "loss": 0.6636, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.4123715771665786e-07, | |
| "loss": 0.6372, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.8257910645009935e-07, | |
| "loss": 0.6312, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.2930371799975593e-07, | |
| "loss": 0.6723, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.814717805502958e-07, | |
| "loss": 0.6655, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.3913787121004717e-07, | |
| "loss": 0.663, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.0235029373752758e-07, | |
| "loss": 0.6555, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 7.115102342598101e-08, | |
| "loss": 0.6702, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 4.55756592088058e-08, | |
| "loss": 0.6347, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.5653383040524228e-08, | |
| "loss": 0.6546, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.1406926599646373e-08, | |
| "loss": 0.6108, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 2.8525453514099966e-09, | |
| "loss": 0.6305, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.5786, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 96, | |
| "total_flos": 6069319925760.0, | |
| "train_loss": 0.9256399211784204, | |
| "train_runtime": 1219.4836, | |
| "train_samples_per_second": 4.92, | |
| "train_steps_per_second": 0.079 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 96, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50000, | |
| "total_flos": 6069319925760.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |