|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.9570093457943925, |
|
"eval_steps": 17, |
|
"global_step": 132, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.601767539978027, |
|
"learning_rate": 1.0000000000000002e-06, |
|
"loss": 2.5188, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"eval_loss": 2.5281636714935303, |
|
"eval_runtime": 136.1276, |
|
"eval_samples_per_second": 6.795, |
|
"eval_steps_per_second": 0.852, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 8.70827865600586, |
|
"learning_rate": 2.0000000000000003e-06, |
|
"loss": 2.561, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 8.504729270935059, |
|
"learning_rate": 3e-06, |
|
"loss": 2.5392, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 8.70252513885498, |
|
"learning_rate": 4.000000000000001e-06, |
|
"loss": 2.5358, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"grad_norm": 8.545235633850098, |
|
"learning_rate": 5e-06, |
|
"loss": 2.4953, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"grad_norm": 8.594281196594238, |
|
"learning_rate": 6e-06, |
|
"loss": 2.5025, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 8.570807456970215, |
|
"learning_rate": 7e-06, |
|
"loss": 2.4483, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"grad_norm": 8.201045036315918, |
|
"learning_rate": 8.000000000000001e-06, |
|
"loss": 2.3844, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"grad_norm": 8.294769287109375, |
|
"learning_rate": 9e-06, |
|
"loss": 2.3132, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 7.825733661651611, |
|
"learning_rate": 1e-05, |
|
"loss": 2.2073, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"grad_norm": 7.359045028686523, |
|
"learning_rate": 9.999301905929286e-06, |
|
"loss": 2.0469, |
|
"step": 11 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"grad_norm": 6.876375198364258, |
|
"learning_rate": 9.997207818651273e-06, |
|
"loss": 1.8699, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"grad_norm": 6.270627021789551, |
|
"learning_rate": 9.99371832291393e-06, |
|
"loss": 1.6649, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"grad_norm": 5.742035388946533, |
|
"learning_rate": 9.988834393115768e-06, |
|
"loss": 1.4954, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 5.249215602874756, |
|
"learning_rate": 9.982557393033758e-06, |
|
"loss": 1.3214, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"grad_norm": 4.818938732147217, |
|
"learning_rate": 9.97488907544252e-06, |
|
"loss": 1.1535, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 4.380295753479004, |
|
"learning_rate": 9.965831581624872e-06, |
|
"loss": 1.0047, |
|
"step": 17 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"eval_loss": 0.8628284335136414, |
|
"eval_runtime": 133.5823, |
|
"eval_samples_per_second": 6.925, |
|
"eval_steps_per_second": 0.868, |
|
"step": 17 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"grad_norm": 3.99696683883667, |
|
"learning_rate": 9.955387440773902e-06, |
|
"loss": 0.8632, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"grad_norm": 3.7503676414489746, |
|
"learning_rate": 9.943559569286731e-06, |
|
"loss": 0.7559, |
|
"step": 19 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 3.430863618850708, |
|
"learning_rate": 9.930351269950144e-06, |
|
"loss": 0.6444, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"grad_norm": 3.25299072265625, |
|
"learning_rate": 9.915766231018317e-06, |
|
"loss": 0.5364, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"grad_norm": 3.217674970626831, |
|
"learning_rate": 9.899808525182935e-06, |
|
"loss": 0.4766, |
|
"step": 22 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"grad_norm": 3.2001354694366455, |
|
"learning_rate": 9.882482608435924e-06, |
|
"loss": 0.4038, |
|
"step": 23 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"grad_norm": 3.011241912841797, |
|
"learning_rate": 9.863793318825186e-06, |
|
"loss": 0.3333, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 2.758089303970337, |
|
"learning_rate": 9.843745875103628e-06, |
|
"loss": 0.2752, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"grad_norm": 1.914292812347412, |
|
"learning_rate": 9.822345875271884e-06, |
|
"loss": 0.2229, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 2.1877121925354004, |
|
"learning_rate": 9.799599295015154e-06, |
|
"loss": 0.1846, |
|
"step": 27 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"grad_norm": 2.3617541790008545, |
|
"learning_rate": 9.775512486034564e-06, |
|
"loss": 0.1556, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 1.431768774986267, |
|
"learning_rate": 9.75009217427352e-06, |
|
"loss": 0.1279, |
|
"step": 29 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 0.7117260098457336, |
|
"learning_rate": 9.723345458039595e-06, |
|
"loss": 0.1061, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 0.45769569277763367, |
|
"learning_rate": 9.695279806022391e-06, |
|
"loss": 0.0987, |
|
"step": 31 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"grad_norm": 0.2953682243824005, |
|
"learning_rate": 9.665903055208013e-06, |
|
"loss": 0.0919, |
|
"step": 32 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 0.17375054955482483, |
|
"learning_rate": 9.635223408690688e-06, |
|
"loss": 0.0854, |
|
"step": 33 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"grad_norm": 0.11500216275453568, |
|
"learning_rate": 9.603249433382145e-06, |
|
"loss": 0.086, |
|
"step": 34 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"eval_loss": 0.08624568581581116, |
|
"eval_runtime": 132.1775, |
|
"eval_samples_per_second": 6.998, |
|
"eval_steps_per_second": 0.878, |
|
"step": 34 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"grad_norm": 0.12646625936031342, |
|
"learning_rate": 9.569990057619414e-06, |
|
"loss": 0.0851, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"grad_norm": 0.11912114918231964, |
|
"learning_rate": 9.535454568671705e-06, |
|
"loss": 0.0834, |
|
"step": 36 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 0.24543708562850952, |
|
"learning_rate": 9.49965261014704e-06, |
|
"loss": 0.0836, |
|
"step": 37 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"grad_norm": 0.0936342254281044, |
|
"learning_rate": 9.462594179299408e-06, |
|
"loss": 0.0844, |
|
"step": 38 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 0.18341362476348877, |
|
"learning_rate": 9.424289624237143e-06, |
|
"loss": 0.0805, |
|
"step": 39 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 0.21815626323223114, |
|
"learning_rate": 9.384749641033358e-06, |
|
"loss": 0.0811, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"grad_norm": 0.21489091217517853, |
|
"learning_rate": 9.343985270739184e-06, |
|
"loss": 0.0793, |
|
"step": 41 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"grad_norm": 0.23281769454479218, |
|
"learning_rate": 9.302007896300697e-06, |
|
"loss": 0.0775, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"grad_norm": 0.6911139488220215, |
|
"learning_rate": 9.25882923938038e-06, |
|
"loss": 0.0812, |
|
"step": 43 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"grad_norm": 0.6087940335273743, |
|
"learning_rate": 9.214461357083986e-06, |
|
"loss": 0.0801, |
|
"step": 44 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"grad_norm": 0.5599693059921265, |
|
"learning_rate": 9.168916638593736e-06, |
|
"loss": 0.0822, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"grad_norm": 0.23726361989974976, |
|
"learning_rate": 9.122207801708802e-06, |
|
"loss": 0.0744, |
|
"step": 46 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 1.075922966003418, |
|
"learning_rate": 9.074347889294017e-06, |
|
"loss": 0.0824, |
|
"step": 47 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"grad_norm": 0.49190425872802734, |
|
"learning_rate": 9.025350265637816e-06, |
|
"loss": 0.0784, |
|
"step": 48 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"grad_norm": 0.8844243288040161, |
|
"learning_rate": 8.975228612720415e-06, |
|
"loss": 0.0722, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"grad_norm": 1.2909235954284668, |
|
"learning_rate": 8.923996926393306e-06, |
|
"loss": 0.075, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"grad_norm": 0.3661656975746155, |
|
"learning_rate": 8.871669512471068e-06, |
|
"loss": 0.0732, |
|
"step": 51 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"eval_loss": 0.07530223578214645, |
|
"eval_runtime": 133.0044, |
|
"eval_samples_per_second": 6.955, |
|
"eval_steps_per_second": 0.872, |
|
"step": 51 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"grad_norm": 0.49425163865089417, |
|
"learning_rate": 8.818260982736662e-06, |
|
"loss": 0.07, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"grad_norm": 0.9367341995239258, |
|
"learning_rate": 8.763786250861258e-06, |
|
"loss": 0.0819, |
|
"step": 53 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"grad_norm": 0.9425927400588989, |
|
"learning_rate": 8.708260528239788e-06, |
|
"loss": 0.0754, |
|
"step": 54 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"grad_norm": 0.3247818052768707, |
|
"learning_rate": 8.651699319743348e-06, |
|
"loss": 0.0739, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"grad_norm": 0.8091337084770203, |
|
"learning_rate": 8.594118419389648e-06, |
|
"loss": 0.0724, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"grad_norm": 0.556105375289917, |
|
"learning_rate": 8.535533905932739e-06, |
|
"loss": 0.0733, |
|
"step": 57 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"grad_norm": 0.9925010800361633, |
|
"learning_rate": 8.475962138373212e-06, |
|
"loss": 0.0761, |
|
"step": 58 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"grad_norm": 0.32729703187942505, |
|
"learning_rate": 8.415419751390155e-06, |
|
"loss": 0.0694, |
|
"step": 59 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"grad_norm": 0.2549174129962921, |
|
"learning_rate": 8.353923650696119e-06, |
|
"loss": 0.0672, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"grad_norm": 0.2688353657722473, |
|
"learning_rate": 8.291491008316409e-06, |
|
"loss": 0.0694, |
|
"step": 61 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"grad_norm": 0.38362765312194824, |
|
"learning_rate": 8.228139257794012e-06, |
|
"loss": 0.0671, |
|
"step": 62 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"grad_norm": 0.8581087589263916, |
|
"learning_rate": 8.163886089321493e-06, |
|
"loss": 0.0745, |
|
"step": 63 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"grad_norm": 0.5643619894981384, |
|
"learning_rate": 8.098749444801226e-06, |
|
"loss": 0.0681, |
|
"step": 64 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"grad_norm": 0.637834906578064, |
|
"learning_rate": 8.032747512835338e-06, |
|
"loss": 0.0773, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"grad_norm": 0.2533693313598633, |
|
"learning_rate": 7.965898723646777e-06, |
|
"loss": 0.0663, |
|
"step": 66 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 0.35648882389068604, |
|
"learning_rate": 7.898221743932887e-06, |
|
"loss": 0.0573, |
|
"step": 67 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"grad_norm": 0.638164758682251, |
|
"learning_rate": 7.829735471652978e-06, |
|
"loss": 0.0719, |
|
"step": 68 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"eval_loss": 0.07532496750354767, |
|
"eval_runtime": 133.6875, |
|
"eval_samples_per_second": 6.919, |
|
"eval_steps_per_second": 0.868, |
|
"step": 68 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"grad_norm": 0.6950666308403015, |
|
"learning_rate": 7.760459030751285e-06, |
|
"loss": 0.0712, |
|
"step": 69 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"grad_norm": 0.5866235494613647, |
|
"learning_rate": 7.690411765816864e-06, |
|
"loss": 0.0753, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"grad_norm": 0.5796181559562683, |
|
"learning_rate": 7.619613236681845e-06, |
|
"loss": 0.0672, |
|
"step": 71 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"grad_norm": 0.5313260555267334, |
|
"learning_rate": 7.548083212959588e-06, |
|
"loss": 0.0634, |
|
"step": 72 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"grad_norm": 0.4373992681503296, |
|
"learning_rate": 7.475841668524268e-06, |
|
"loss": 0.0632, |
|
"step": 73 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"grad_norm": 0.3541436195373535, |
|
"learning_rate": 7.402908775933419e-06, |
|
"loss": 0.0675, |
|
"step": 74 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"grad_norm": 0.5151358842849731, |
|
"learning_rate": 7.329304900794991e-06, |
|
"loss": 0.066, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"grad_norm": 0.6343135833740234, |
|
"learning_rate": 7.25505059608051e-06, |
|
"loss": 0.0676, |
|
"step": 76 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"grad_norm": 0.6041606664657593, |
|
"learning_rate": 7.180166596385915e-06, |
|
"loss": 0.0659, |
|
"step": 77 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"grad_norm": 0.5856900811195374, |
|
"learning_rate": 7.104673812141676e-06, |
|
"loss": 0.0585, |
|
"step": 78 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"grad_norm": 0.2674119770526886, |
|
"learning_rate": 7.028593323773819e-06, |
|
"loss": 0.0688, |
|
"step": 79 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"grad_norm": 0.5323411226272583, |
|
"learning_rate": 6.9519463758174745e-06, |
|
"loss": 0.0645, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"grad_norm": 0.5299087166786194, |
|
"learning_rate": 6.8747543709846064e-06, |
|
"loss": 0.0631, |
|
"step": 81 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"grad_norm": 1.1804003715515137, |
|
"learning_rate": 6.797038864187564e-06, |
|
"loss": 0.0813, |
|
"step": 82 |
|
}, |
|
{ |
|
"epoch": 1.22, |
|
"grad_norm": 1.129088044166565, |
|
"learning_rate": 6.718821556520151e-06, |
|
"loss": 0.0661, |
|
"step": 83 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"grad_norm": 0.37926897406578064, |
|
"learning_rate": 6.640124289197845e-06, |
|
"loss": 0.0739, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"grad_norm": 0.5506991744041443, |
|
"learning_rate": 6.560969037458933e-06, |
|
"loss": 0.0722, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"eval_loss": 0.06798132508993149, |
|
"eval_runtime": 131.6498, |
|
"eval_samples_per_second": 7.026, |
|
"eval_steps_per_second": 0.881, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 1.27, |
|
"grad_norm": 0.36365798115730286, |
|
"learning_rate": 6.481377904428171e-06, |
|
"loss": 0.0675, |
|
"step": 86 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"grad_norm": 0.24918232858181, |
|
"learning_rate": 6.401373114944781e-06, |
|
"loss": 0.0654, |
|
"step": 87 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"grad_norm": 0.32376477122306824, |
|
"learning_rate": 6.3209770093564315e-06, |
|
"loss": 0.0666, |
|
"step": 88 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"grad_norm": 0.5324047207832336, |
|
"learning_rate": 6.240212037280967e-06, |
|
"loss": 0.0616, |
|
"step": 89 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"grad_norm": 0.2454373985528946, |
|
"learning_rate": 6.1591007513376425e-06, |
|
"loss": 0.0705, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"grad_norm": 0.5578957796096802, |
|
"learning_rate": 6.077665800849568e-06, |
|
"loss": 0.0783, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"grad_norm": 0.475643515586853, |
|
"learning_rate": 5.995929925519181e-06, |
|
"loss": 0.0689, |
|
"step": 92 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"grad_norm": 0.5447214841842651, |
|
"learning_rate": 5.913915949078453e-06, |
|
"loss": 0.0692, |
|
"step": 93 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"grad_norm": 0.19609016180038452, |
|
"learning_rate": 5.831646772915651e-06, |
|
"loss": 0.0684, |
|
"step": 94 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"grad_norm": 0.3513399362564087, |
|
"learning_rate": 5.7491453696804075e-06, |
|
"loss": 0.0606, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"grad_norm": 0.5132963061332703, |
|
"learning_rate": 5.666434776868895e-06, |
|
"loss": 0.0618, |
|
"step": 96 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"grad_norm": 0.3573090732097626, |
|
"learning_rate": 5.583538090390882e-06, |
|
"loss": 0.066, |
|
"step": 97 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"grad_norm": 0.37191343307495117, |
|
"learning_rate": 5.500478458120493e-06, |
|
"loss": 0.0574, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"grad_norm": 0.5762889385223389, |
|
"learning_rate": 5.41727907343245e-06, |
|
"loss": 0.0698, |
|
"step": 99 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"grad_norm": 0.2511120140552521, |
|
"learning_rate": 5.3339631687256085e-06, |
|
"loss": 0.0642, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"grad_norm": 0.5286504626274109, |
|
"learning_rate": 5.250554008935596e-06, |
|
"loss": 0.0702, |
|
"step": 101 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"grad_norm": 0.283104807138443, |
|
"learning_rate": 5.1670748850383734e-06, |
|
"loss": 0.0676, |
|
"step": 102 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"eval_loss": 0.06662092357873917, |
|
"eval_runtime": 128.4351, |
|
"eval_samples_per_second": 7.202, |
|
"eval_steps_per_second": 0.903, |
|
"step": 102 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"grad_norm": 0.3505866527557373, |
|
"learning_rate": 5.083549107546505e-06, |
|
"loss": 0.0749, |
|
"step": 103 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"grad_norm": 0.4083852469921112, |
|
"learning_rate": 5e-06, |
|
"loss": 0.0683, |
|
"step": 104 |
|
}, |
|
{ |
|
"epoch": 1.55, |
|
"grad_norm": 0.6204113364219666, |
|
"learning_rate": 4.916450892453495e-06, |
|
"loss": 0.0579, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 0.2663183808326721, |
|
"learning_rate": 4.832925114961629e-06, |
|
"loss": 0.0606, |
|
"step": 106 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"grad_norm": 0.5618534684181213, |
|
"learning_rate": 4.7494459910644044e-06, |
|
"loss": 0.0702, |
|
"step": 107 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 0.5477017760276794, |
|
"learning_rate": 4.666036831274392e-06, |
|
"loss": 0.0717, |
|
"step": 108 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"grad_norm": 0.5102267265319824, |
|
"learning_rate": 4.582720926567552e-06, |
|
"loss": 0.0703, |
|
"step": 109 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"grad_norm": 0.2536521255970001, |
|
"learning_rate": 4.499521541879508e-06, |
|
"loss": 0.0653, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"grad_norm": 0.3182464838027954, |
|
"learning_rate": 4.416461909609119e-06, |
|
"loss": 0.068, |
|
"step": 111 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"grad_norm": 0.2570602595806122, |
|
"learning_rate": 4.333565223131107e-06, |
|
"loss": 0.0636, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"grad_norm": 0.1940167397260666, |
|
"learning_rate": 4.250854630319593e-06, |
|
"loss": 0.0692, |
|
"step": 113 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"grad_norm": 0.33302024006843567, |
|
"learning_rate": 4.1683532270843505e-06, |
|
"loss": 0.062, |
|
"step": 114 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"grad_norm": 0.26795655488967896, |
|
"learning_rate": 4.08608405092155e-06, |
|
"loss": 0.0629, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"grad_norm": 0.5258805751800537, |
|
"learning_rate": 4.004070074480821e-06, |
|
"loss": 0.0634, |
|
"step": 116 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"grad_norm": 0.2478051334619522, |
|
"learning_rate": 3.922334199150433e-06, |
|
"loss": 0.0675, |
|
"step": 117 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"grad_norm": 0.30332913994789124, |
|
"learning_rate": 3.840899248662358e-06, |
|
"loss": 0.063, |
|
"step": 118 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"grad_norm": 0.21814480423927307, |
|
"learning_rate": 3.7597879627190337e-06, |
|
"loss": 0.068, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"eval_loss": 0.06482138484716415, |
|
"eval_runtime": 126.2918, |
|
"eval_samples_per_second": 7.324, |
|
"eval_steps_per_second": 0.919, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"grad_norm": 0.27936533093452454, |
|
"learning_rate": 3.6790229906435706e-06, |
|
"loss": 0.065, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"grad_norm": 0.2364480048418045, |
|
"learning_rate": 3.598626885055219e-06, |
|
"loss": 0.0672, |
|
"step": 121 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"grad_norm": 0.32356905937194824, |
|
"learning_rate": 3.518622095571831e-06, |
|
"loss": 0.065, |
|
"step": 122 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"grad_norm": 0.24160273373126984, |
|
"learning_rate": 3.439030962541069e-06, |
|
"loss": 0.0646, |
|
"step": 123 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"grad_norm": 0.26129576563835144, |
|
"learning_rate": 3.3598757108021546e-06, |
|
"loss": 0.064, |
|
"step": 124 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"grad_norm": 0.24441897869110107, |
|
"learning_rate": 3.281178443479852e-06, |
|
"loss": 0.0525, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"grad_norm": 0.23377354443073273, |
|
"learning_rate": 3.202961135812437e-06, |
|
"loss": 0.0592, |
|
"step": 126 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"grad_norm": 0.3297436833381653, |
|
"learning_rate": 3.1252456290153952e-06, |
|
"loss": 0.0625, |
|
"step": 127 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"grad_norm": 0.7371227145195007, |
|
"learning_rate": 3.0480536241825263e-06, |
|
"loss": 0.0644, |
|
"step": 128 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"grad_norm": 0.46529266238212585, |
|
"learning_rate": 2.9714066762261825e-06, |
|
"loss": 0.0641, |
|
"step": 129 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"grad_norm": 0.5547051429748535, |
|
"learning_rate": 2.8953261878583263e-06, |
|
"loss": 0.0747, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"grad_norm": 0.3253116011619568, |
|
"learning_rate": 2.8198334036140873e-06, |
|
"loss": 0.0709, |
|
"step": 131 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"grad_norm": 0.26438066363334656, |
|
"learning_rate": 2.74494940391949e-06, |
|
"loss": 0.0651, |
|
"step": 132 |
|
} |
|
], |
|
"logging_steps": 1, |
|
"max_steps": 198, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 66, |
|
"total_flos": 9.335735208168653e+16, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|