| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0021892057561516682, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.9189410082022242e-05, | |
| "grad_norm": 6.656045436859131, | |
| "learning_rate": 1e-05, | |
| "loss": 6.6322, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.9189410082022242e-05, | |
| "eval_loss": 4.265627384185791, | |
| "eval_runtime": 5138.7283, | |
| "eval_samples_per_second": 5.614, | |
| "eval_steps_per_second": 0.702, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.8378820164044484e-05, | |
| "grad_norm": 9.15876579284668, | |
| "learning_rate": 2e-05, | |
| "loss": 9.203, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 8.756823024606673e-05, | |
| "grad_norm": 7.754046440124512, | |
| "learning_rate": 3e-05, | |
| "loss": 7.807, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00011675764032808897, | |
| "grad_norm": 8.308610916137695, | |
| "learning_rate": 4e-05, | |
| "loss": 8.9153, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0001459470504101112, | |
| "grad_norm": 8.078652381896973, | |
| "learning_rate": 5e-05, | |
| "loss": 8.2432, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00017513646049213346, | |
| "grad_norm": 10.330500602722168, | |
| "learning_rate": 6e-05, | |
| "loss": 9.0539, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0002043258705741557, | |
| "grad_norm": 9.112791061401367, | |
| "learning_rate": 7e-05, | |
| "loss": 7.2966, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00023351528065617793, | |
| "grad_norm": 8.946612358093262, | |
| "learning_rate": 8e-05, | |
| "loss": 6.4122, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0002627046907382002, | |
| "grad_norm": 7.859531879425049, | |
| "learning_rate": 9e-05, | |
| "loss": 6.1909, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0002627046907382002, | |
| "eval_loss": 2.7228097915649414, | |
| "eval_runtime": 5145.5769, | |
| "eval_samples_per_second": 5.607, | |
| "eval_steps_per_second": 0.701, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0002918941008202224, | |
| "grad_norm": 8.455995559692383, | |
| "learning_rate": 0.0001, | |
| "loss": 5.3154, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00032108351090224467, | |
| "grad_norm": 6.980202674865723, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 5.7249, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00035027292098426693, | |
| "grad_norm": 10.977880477905273, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.3159, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00037946233106628914, | |
| "grad_norm": 5.031854629516602, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 3.0523, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0004086517411483114, | |
| "grad_norm": 5.270071029663086, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.4981, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00043784115123033366, | |
| "grad_norm": 4.871089458465576, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.3594, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00046703056131235587, | |
| "grad_norm": 6.282345771789551, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.8028, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0004962199713943781, | |
| "grad_norm": 3.952486991882324, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.8249, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0005254093814764004, | |
| "grad_norm": 5.634269714355469, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.0929, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0005254093814764004, | |
| "eval_loss": 1.4089776277542114, | |
| "eval_runtime": 5146.9482, | |
| "eval_samples_per_second": 5.605, | |
| "eval_steps_per_second": 0.701, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0005545987915584227, | |
| "grad_norm": 4.055173397064209, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.4124, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0005837882016404448, | |
| "grad_norm": 6.300996780395508, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.6939, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006129776117224671, | |
| "grad_norm": 4.606357574462891, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.2878, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0006421670218044893, | |
| "grad_norm": 4.433748245239258, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.2333, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0006713564318865116, | |
| "grad_norm": 5.106141567230225, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.9466, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0007005458419685339, | |
| "grad_norm": 4.833502769470215, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.3432, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.000729735252050556, | |
| "grad_norm": 5.314240455627441, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.9879, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0007589246621325783, | |
| "grad_norm": 6.109968662261963, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 2.9088, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0007881140722146005, | |
| "grad_norm": 4.610195159912109, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.0524, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0007881140722146005, | |
| "eval_loss": 1.2951326370239258, | |
| "eval_runtime": 5145.9741, | |
| "eval_samples_per_second": 5.606, | |
| "eval_steps_per_second": 0.701, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0008173034822966228, | |
| "grad_norm": 5.078436851501465, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.9604, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0008464928923786451, | |
| "grad_norm": 7.848934650421143, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.8579, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0008756823024606673, | |
| "grad_norm": 5.513472080230713, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.5474, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0009048717125426895, | |
| "grad_norm": 5.696840286254883, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.6489, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0009340611226247117, | |
| "grad_norm": 3.9993512630462646, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.9396, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.000963250532706734, | |
| "grad_norm": 5.254062652587891, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.6952, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0009924399427887563, | |
| "grad_norm": 5.616064071655273, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.5141, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0010216293528707784, | |
| "grad_norm": 5.979726791381836, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.7039, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0010508187629528008, | |
| "grad_norm": 4.942209243774414, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.9582, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0010508187629528008, | |
| "eval_loss": 1.2488832473754883, | |
| "eval_runtime": 5146.2141, | |
| "eval_samples_per_second": 5.606, | |
| "eval_steps_per_second": 0.701, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.001080008173034823, | |
| "grad_norm": 4.169759273529053, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.0918, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0011091975831168453, | |
| "grad_norm": 4.325538158416748, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 2.4416, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0011383869931988675, | |
| "grad_norm": 5.6241912841796875, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.076, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0011675764032808896, | |
| "grad_norm": 3.756821870803833, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.2152, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001196765813362912, | |
| "grad_norm": 4.458866119384766, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 3.2741, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0012259552234449341, | |
| "grad_norm": 3.5846951007843018, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.208, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0012551446335269565, | |
| "grad_norm": 5.011172294616699, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.823, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0012843340436089787, | |
| "grad_norm": 3.8864340782165527, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.9924, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0013135234536910008, | |
| "grad_norm": 4.546591281890869, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.8482, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0013135234536910008, | |
| "eval_loss": 1.2012851238250732, | |
| "eval_runtime": 5145.245, | |
| "eval_samples_per_second": 5.607, | |
| "eval_steps_per_second": 0.701, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0013427128637730232, | |
| "grad_norm": 4.4162678718566895, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.1812, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0013719022738550453, | |
| "grad_norm": 4.07814359664917, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 2.4973, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0014010916839370677, | |
| "grad_norm": 5.748778820037842, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.4652, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0014302810940190899, | |
| "grad_norm": 4.544031143188477, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.6362, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.001459470504101112, | |
| "grad_norm": 4.541274070739746, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.4015, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0014886599141831344, | |
| "grad_norm": 4.629258155822754, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.8155, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0015178493242651565, | |
| "grad_norm": 3.993497371673584, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2.2827, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.001547038734347179, | |
| "grad_norm": 4.616318702697754, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 2.6136, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.001576228144429201, | |
| "grad_norm": 3.513707399368286, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.0149, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.001576228144429201, | |
| "eval_loss": 1.1747256517410278, | |
| "eval_runtime": 5143.0061, | |
| "eval_samples_per_second": 5.61, | |
| "eval_steps_per_second": 0.701, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0016054175545112232, | |
| "grad_norm": 3.937657356262207, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0302, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0016346069645932456, | |
| "grad_norm": 5.150801658630371, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 2.5411, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.0016637963746752677, | |
| "grad_norm": 4.555789470672607, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.8977, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0016929857847572901, | |
| "grad_norm": 3.8810622692108154, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 2.0767, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.0017221751948393123, | |
| "grad_norm": 4.165887832641602, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 2.3894, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.0017513646049213346, | |
| "grad_norm": 3.750863552093506, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.0504, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0017805540150033568, | |
| "grad_norm": 2.934596538543701, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 1.9172, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.001809743425085379, | |
| "grad_norm": 3.5289196968078613, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 2.2315, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.0018389328351674013, | |
| "grad_norm": 4.346237659454346, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.2884, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0018389328351674013, | |
| "eval_loss": 1.147663950920105, | |
| "eval_runtime": 5144.0405, | |
| "eval_samples_per_second": 5.608, | |
| "eval_steps_per_second": 0.701, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0018681222452494235, | |
| "grad_norm": 5.116072177886963, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.4416, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.0018973116553314458, | |
| "grad_norm": 3.6044180393218994, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 2.6073, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.001926501065413468, | |
| "grad_norm": 3.5492379665374756, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.714, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.00195569047549549, | |
| "grad_norm": 3.5761613845825195, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 1.8829, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.0019848798855775125, | |
| "grad_norm": 3.665513515472412, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 2.003, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.002014069295659535, | |
| "grad_norm": 4.549952030181885, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.0985, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.002043258705741557, | |
| "grad_norm": 6.244661808013916, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 2.3304, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.002072448115823579, | |
| "grad_norm": 4.266684055328369, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 2.0038, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.0021016375259056016, | |
| "grad_norm": 4.135928153991699, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.1593, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0021016375259056016, | |
| "eval_loss": 1.1359862089157104, | |
| "eval_runtime": 5144.5001, | |
| "eval_samples_per_second": 5.608, | |
| "eval_steps_per_second": 0.701, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0021308269359876235, | |
| "grad_norm": 4.030326843261719, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.9979, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.002160016346069646, | |
| "grad_norm": 4.288099765777588, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 2.3285, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.0021892057561516682, | |
| "grad_norm": 3.925044536590576, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.015, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1108631969792e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |