| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05951200158698671, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007934933544931561, | |
| "grad_norm": 48.7606315612793, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 29.6105, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007934933544931561, | |
| "eval_loss": 3.7225465774536133, | |
| "eval_runtime": 212.1774, | |
| "eval_samples_per_second": 5.005, | |
| "eval_steps_per_second": 2.503, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0015869867089863122, | |
| "grad_norm": 46.967803955078125, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 29.9806, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0023804800634794686, | |
| "grad_norm": 50.5672492980957, | |
| "learning_rate": 0.0002, | |
| "loss": 26.9378, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0031739734179726245, | |
| "grad_norm": 46.023780822753906, | |
| "learning_rate": 0.0001999048221581858, | |
| "loss": 21.9639, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003967466772465781, | |
| "grad_norm": 31.295907974243164, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 10.6674, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004760960126958937, | |
| "grad_norm": 23.777172088623047, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 6.2121, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005554453481452093, | |
| "grad_norm": 23.954105377197266, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 4.3349, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006347946835945249, | |
| "grad_norm": 7.861544132232666, | |
| "learning_rate": 0.00019762960071199333, | |
| "loss": 3.1486, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007141440190438405, | |
| "grad_norm": 6.988774299621582, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 2.6602, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007934933544931562, | |
| "grad_norm": 32.20113754272461, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 4.0596, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008728426899424718, | |
| "grad_norm": 14.095051765441895, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 2.995, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009521920253917874, | |
| "grad_norm": 13.5059232711792, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 3.2571, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010315413608411029, | |
| "grad_norm": 7.761572360992432, | |
| "learning_rate": 0.000190630778703665, | |
| "loss": 3.0002, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011108906962904185, | |
| "grad_norm": 4.439308166503906, | |
| "learning_rate": 0.00018870108331782217, | |
| "loss": 2.6362, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011902400317397342, | |
| "grad_norm": 6.415318489074707, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 3.1925, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012695893671890498, | |
| "grad_norm": 6.409482955932617, | |
| "learning_rate": 0.0001843391445812886, | |
| "loss": 3.062, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013489387026383654, | |
| "grad_norm": 3.2289886474609375, | |
| "learning_rate": 0.0001819152044288992, | |
| "loss": 2.8807, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01428288038087681, | |
| "grad_norm": 7.170831680297852, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 3.1952, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.015076373735369967, | |
| "grad_norm": 8.997602462768555, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 2.9534, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.015869867089863123, | |
| "grad_norm": 18.465681076049805, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 6.0253, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01666336044435628, | |
| "grad_norm": 6.664717674255371, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.1157, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017456853798849436, | |
| "grad_norm": 6.979747772216797, | |
| "learning_rate": 0.00016755902076156604, | |
| "loss": 3.0816, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.018250347153342592, | |
| "grad_norm": 4.373898506164551, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 2.921, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.01904384050783575, | |
| "grad_norm": 3.7028324604034424, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 2.8931, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0198373338623289, | |
| "grad_norm": 2.8346545696258545, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 2.8573, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0198373338623289, | |
| "eval_loss": 0.37545695900917053, | |
| "eval_runtime": 213.9659, | |
| "eval_samples_per_second": 4.963, | |
| "eval_steps_per_second": 2.482, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.020630827216822058, | |
| "grad_norm": 3.2634661197662354, | |
| "learning_rate": 0.0001537299608346824, | |
| "loss": 2.9078, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021424320571315214, | |
| "grad_norm": 4.43280553817749, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.0779, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.02221781392580837, | |
| "grad_norm": 3.53800106048584, | |
| "learning_rate": 0.00014617486132350343, | |
| "loss": 2.9961, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.023011307280301527, | |
| "grad_norm": 2.2324697971343994, | |
| "learning_rate": 0.00014226182617406996, | |
| "loss": 2.7108, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.023804800634794683, | |
| "grad_norm": 3.370065212249756, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 2.9733, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02459829398928784, | |
| "grad_norm": 2.1509768962860107, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 2.7909, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.025391787343780996, | |
| "grad_norm": 3.518296957015991, | |
| "learning_rate": 0.00013007057995042732, | |
| "loss": 3.0504, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.026185280698274152, | |
| "grad_norm": 2.118720054626465, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 2.8059, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02697877405276731, | |
| "grad_norm": 2.4879894256591797, | |
| "learning_rate": 0.00012164396139381029, | |
| "loss": 2.8901, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.027772267407260465, | |
| "grad_norm": 2.6533515453338623, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 2.8234, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02856576076175362, | |
| "grad_norm": 8.765356063842773, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 4.3363, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.029359254116246777, | |
| "grad_norm": 2.3454673290252686, | |
| "learning_rate": 0.00010871557427476583, | |
| "loss": 3.0552, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.030152747470739934, | |
| "grad_norm": 1.8576560020446777, | |
| "learning_rate": 0.00010436193873653361, | |
| "loss": 2.8869, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.03094624082523309, | |
| "grad_norm": 2.4082350730895996, | |
| "learning_rate": 0.0001, | |
| "loss": 2.8681, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.031739734179726246, | |
| "grad_norm": 2.3808937072753906, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 2.8245, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0325332275342194, | |
| "grad_norm": 1.8626981973648071, | |
| "learning_rate": 9.128442572523417e-05, | |
| "loss": 2.8774, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03332672088871256, | |
| "grad_norm": 2.4850103855133057, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 2.9511, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.034120214243205715, | |
| "grad_norm": 2.5530030727386475, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 2.7849, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03491370759769887, | |
| "grad_norm": 3.354628562927246, | |
| "learning_rate": 7.835603860618972e-05, | |
| "loss": 2.8894, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03570720095219203, | |
| "grad_norm": 2.490819215774536, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 2.8177, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.036500694306685184, | |
| "grad_norm": 4.005395412445068, | |
| "learning_rate": 6.992942004957271e-05, | |
| "loss": 3.0671, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03729418766117834, | |
| "grad_norm": 2.593477725982666, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 2.9372, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0380876810156715, | |
| "grad_norm": 4.0660319328308105, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 2.9659, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.038881174370164646, | |
| "grad_norm": 3.1026699542999268, | |
| "learning_rate": 5.773817382593008e-05, | |
| "loss": 3.0151, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0396746677246578, | |
| "grad_norm": 6.865238189697266, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 2.9836, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0396746677246578, | |
| "eval_loss": 0.3641993999481201, | |
| "eval_runtime": 214.9129, | |
| "eval_samples_per_second": 4.942, | |
| "eval_steps_per_second": 2.471, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04046816107915096, | |
| "grad_norm": 3.206982374191284, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 2.8024, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.041261654433644115, | |
| "grad_norm": 2.6523642539978027, | |
| "learning_rate": 4.6270039165317605e-05, | |
| "loss": 2.9125, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.04205514778813727, | |
| "grad_norm": 2.145852565765381, | |
| "learning_rate": 4.264235636489542e-05, | |
| "loss": 2.7628, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.04284864114263043, | |
| "grad_norm": 1.956112027168274, | |
| "learning_rate": 3.9123857099127936e-05, | |
| "loss": 2.829, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.043642134497123584, | |
| "grad_norm": 3.5577127933502197, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 2.9393, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.04443562785161674, | |
| "grad_norm": 2.462873697280884, | |
| "learning_rate": 3.244097923843398e-05, | |
| "loss": 2.9588, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.0452291212061099, | |
| "grad_norm": 1.5341908931732178, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 2.7976, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.04602261456060305, | |
| "grad_norm": 3.1381309032440186, | |
| "learning_rate": 2.6272266318987603e-05, | |
| "loss": 2.8753, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.04681610791509621, | |
| "grad_norm": 3.1369926929473877, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 2.8046, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.047609601269589366, | |
| "grad_norm": 3.5989041328430176, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 3.0486, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04840309462408252, | |
| "grad_norm": 3.038649559020996, | |
| "learning_rate": 1.808479557110081e-05, | |
| "loss": 2.8922, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.04919658797857568, | |
| "grad_norm": 3.415806293487549, | |
| "learning_rate": 1.566085541871145e-05, | |
| "loss": 2.7841, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.049990081333068835, | |
| "grad_norm": 1.7945884466171265, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 2.9487, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.05078357468756199, | |
| "grad_norm": 2.6868271827697754, | |
| "learning_rate": 1.129891668217783e-05, | |
| "loss": 2.8241, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.05157706804205515, | |
| "grad_norm": 2.454937219619751, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 2.7385, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.052370561396548304, | |
| "grad_norm": 2.422858953475952, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 2.9928, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.05316405475104146, | |
| "grad_norm": 2.9807822704315186, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 2.873, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.05395754810553462, | |
| "grad_norm": 2.892643451690674, | |
| "learning_rate": 4.628304925177318e-06, | |
| "loss": 2.8405, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.05475104146002777, | |
| "grad_norm": 2.1034739017486572, | |
| "learning_rate": 3.40741737109318e-06, | |
| "loss": 2.7657, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.05554453481452093, | |
| "grad_norm": 2.7781882286071777, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 2.9382, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.056338028169014086, | |
| "grad_norm": 1.8283110857009888, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 2.809, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.05713152152350724, | |
| "grad_norm": 3.2760870456695557, | |
| "learning_rate": 8.555138626189618e-07, | |
| "loss": 2.9855, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0579250148780004, | |
| "grad_norm": 2.1711838245391846, | |
| "learning_rate": 3.805301908254455e-07, | |
| "loss": 2.914, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.058718508232493555, | |
| "grad_norm": 1.9169692993164062, | |
| "learning_rate": 9.517784181422019e-08, | |
| "loss": 2.868, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.05951200158698671, | |
| "grad_norm": 2.8653454780578613, | |
| "learning_rate": 0.0, | |
| "loss": 2.9482, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05951200158698671, | |
| "eval_loss": 0.36626261472702026, | |
| "eval_runtime": 214.2761, | |
| "eval_samples_per_second": 4.956, | |
| "eval_steps_per_second": 2.478, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.80187302854656e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |