| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0396746677246578, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007934933544931561, | |
| "grad_norm": 50.82598876953125, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 29.6105, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007934933544931561, | |
| "eval_loss": 3.7225465774536133, | |
| "eval_runtime": 210.8547, | |
| "eval_samples_per_second": 5.037, | |
| "eval_steps_per_second": 2.518, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0015869867089863122, | |
| "grad_norm": 48.4813232421875, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 29.9806, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0023804800634794686, | |
| "grad_norm": 51.97000503540039, | |
| "learning_rate": 0.0002, | |
| "loss": 26.8948, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0031739734179726245, | |
| "grad_norm": 46.51694869995117, | |
| "learning_rate": 0.0001999048221581858, | |
| "loss": 21.9835, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003967466772465781, | |
| "grad_norm": 31.5517635345459, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 10.75, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004760960126958937, | |
| "grad_norm": 23.97246742248535, | |
| "learning_rate": 0.00019914448613738106, | |
| "loss": 6.3342, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005554453481452093, | |
| "grad_norm": 22.029800415039062, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 4.2752, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006347946835945249, | |
| "grad_norm": 8.115840911865234, | |
| "learning_rate": 0.00019762960071199333, | |
| "loss": 3.1482, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007141440190438405, | |
| "grad_norm": 7.256277561187744, | |
| "learning_rate": 0.00019659258262890683, | |
| "loss": 2.6701, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007934933544931562, | |
| "grad_norm": 34.10689926147461, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 4.1156, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008728426899424718, | |
| "grad_norm": 12.535100936889648, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 2.9717, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009521920253917874, | |
| "grad_norm": 13.594742774963379, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 3.2695, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010315413608411029, | |
| "grad_norm": 8.208749771118164, | |
| "learning_rate": 0.000190630778703665, | |
| "loss": 3.0809, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011108906962904185, | |
| "grad_norm": 4.1398773193359375, | |
| "learning_rate": 0.00018870108331782217, | |
| "loss": 2.6611, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011902400317397342, | |
| "grad_norm": 6.833378791809082, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 3.1422, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012695893671890498, | |
| "grad_norm": 6.5985612869262695, | |
| "learning_rate": 0.0001843391445812886, | |
| "loss": 3.1404, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013489387026383654, | |
| "grad_norm": 2.9224817752838135, | |
| "learning_rate": 0.0001819152044288992, | |
| "loss": 2.8423, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01428288038087681, | |
| "grad_norm": 6.999931335449219, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 3.2225, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.015076373735369967, | |
| "grad_norm": 3.803568124771118, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 2.888, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.015869867089863123, | |
| "grad_norm": 15.167760848999023, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 6.0032, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01666336044435628, | |
| "grad_norm": 4.701049327850342, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 3.2004, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017456853798849436, | |
| "grad_norm": 5.967288970947266, | |
| "learning_rate": 0.00016755902076156604, | |
| "loss": 2.896, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.018250347153342592, | |
| "grad_norm": 4.388175010681152, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 2.888, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.01904384050783575, | |
| "grad_norm": 3.3615219593048096, | |
| "learning_rate": 0.00016087614290087208, | |
| "loss": 2.8122, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0198373338623289, | |
| "grad_norm": 3.9244184494018555, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 2.9292, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0198373338623289, | |
| "eval_loss": 0.3786240220069885, | |
| "eval_runtime": 213.3259, | |
| "eval_samples_per_second": 4.978, | |
| "eval_steps_per_second": 2.489, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.020630827216822058, | |
| "grad_norm": 3.8264613151550293, | |
| "learning_rate": 0.0001537299608346824, | |
| "loss": 2.9965, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021424320571315214, | |
| "grad_norm": 4.253182411193848, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.9774, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.02221781392580837, | |
| "grad_norm": 3.146336555480957, | |
| "learning_rate": 0.00014617486132350343, | |
| "loss": 2.9017, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.023011307280301527, | |
| "grad_norm": 2.462265968322754, | |
| "learning_rate": 0.00014226182617406996, | |
| "loss": 2.7964, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.023804800634794683, | |
| "grad_norm": 4.6121673583984375, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 3.0148, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02459829398928784, | |
| "grad_norm": 2.4022810459136963, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 2.7315, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.025391787343780996, | |
| "grad_norm": 12.726351737976074, | |
| "learning_rate": 0.00013007057995042732, | |
| "loss": 3.0036, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.026185280698274152, | |
| "grad_norm": 2.4199745655059814, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 2.7967, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02697877405276731, | |
| "grad_norm": 4.458526134490967, | |
| "learning_rate": 0.00012164396139381029, | |
| "loss": 2.9338, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.027772267407260465, | |
| "grad_norm": 4.283806324005127, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 2.8298, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02856576076175362, | |
| "grad_norm": 25.05265998840332, | |
| "learning_rate": 0.00011305261922200519, | |
| "loss": 4.3322, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.029359254116246777, | |
| "grad_norm": 9.16774845123291, | |
| "learning_rate": 0.00010871557427476583, | |
| "loss": 3.1166, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.030152747470739934, | |
| "grad_norm": 9.907901763916016, | |
| "learning_rate": 0.00010436193873653361, | |
| "loss": 2.8739, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.03094624082523309, | |
| "grad_norm": 5.029708385467529, | |
| "learning_rate": 0.0001, | |
| "loss": 2.8051, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.031739734179726246, | |
| "grad_norm": 3.865046739578247, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 2.765, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0325332275342194, | |
| "grad_norm": 3.3393161296844482, | |
| "learning_rate": 9.128442572523417e-05, | |
| "loss": 2.8003, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03332672088871256, | |
| "grad_norm": 2.8670387268066406, | |
| "learning_rate": 8.694738077799488e-05, | |
| "loss": 2.8394, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.034120214243205715, | |
| "grad_norm": 3.05271053314209, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 2.762, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03491370759769887, | |
| "grad_norm": 5.7023024559021, | |
| "learning_rate": 7.835603860618972e-05, | |
| "loss": 2.8546, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03570720095219203, | |
| "grad_norm": 6.604750633239746, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 2.7546, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.036500694306685184, | |
| "grad_norm": 8.486159324645996, | |
| "learning_rate": 6.992942004957271e-05, | |
| "loss": 3.0494, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03729418766117834, | |
| "grad_norm": 4.845094203948975, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 2.9887, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0380876810156715, | |
| "grad_norm": 4.348211288452148, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 2.9466, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.038881174370164646, | |
| "grad_norm": 4.239773750305176, | |
| "learning_rate": 5.773817382593008e-05, | |
| "loss": 3.1212, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0396746677246578, | |
| "grad_norm": 8.90642261505127, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 2.9847, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0396746677246578, | |
| "eval_loss": 0.36671048402786255, | |
| "eval_runtime": 213.134, | |
| "eval_samples_per_second": 4.983, | |
| "eval_steps_per_second": 2.491, | |
| "step": 50 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.53458201903104e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |