| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.5842293906810037, | |
| "eval_steps": 1000, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005973715651135006, | |
| "grad_norm": 8.9375, | |
| "learning_rate": 2e-06, | |
| "loss": 1.3168, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05973715651135006, | |
| "grad_norm": 0.625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5498, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11947431302270012, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3884, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17921146953405018, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0006, | |
| "loss": 0.376, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.23894862604540024, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.0008, | |
| "loss": 0.3635, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2986857825567503, | |
| "grad_norm": 0.1826171875, | |
| "learning_rate": 0.001, | |
| "loss": 0.3559, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.35842293906810035, | |
| "grad_norm": 0.10888671875, | |
| "learning_rate": 0.0012, | |
| "loss": 0.3438, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.41816009557945044, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.0014, | |
| "loss": 0.3412, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.4778972520908005, | |
| "grad_norm": 0.08984375, | |
| "learning_rate": 0.0016, | |
| "loss": 0.3392, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5376344086021505, | |
| "grad_norm": 0.11376953125, | |
| "learning_rate": 0.0018000000000000002, | |
| "loss": 0.346, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5973715651135006, | |
| "grad_norm": 0.08251953125, | |
| "learning_rate": 0.002, | |
| "loss": 0.3174, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6571087216248507, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 0.0019998292504580526, | |
| "loss": 0.3309, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7168458781362007, | |
| "grad_norm": 0.1533203125, | |
| "learning_rate": 0.001999317060143023, | |
| "loss": 0.323, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7765830346475507, | |
| "grad_norm": 0.09228515625, | |
| "learning_rate": 0.001998463603967434, | |
| "loss": 0.3157, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.8363201911589009, | |
| "grad_norm": 0.072265625, | |
| "learning_rate": 0.0019972691733857882, | |
| "loss": 0.3155, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8960573476702509, | |
| "grad_norm": 0.12890625, | |
| "learning_rate": 0.0019957341762950344, | |
| "loss": 0.3148, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.955794504181601, | |
| "grad_norm": 0.10205078125, | |
| "learning_rate": 0.001993859136895274, | |
| "loss": 0.3013, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.015531660692951, | |
| "grad_norm": 0.11474609375, | |
| "learning_rate": 0.0019916446955107426, | |
| "loss": 0.3009, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.075268817204301, | |
| "grad_norm": 0.12890625, | |
| "learning_rate": 0.001989091608371146, | |
| "loss": 0.3005, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.135005973715651, | |
| "grad_norm": 0.087890625, | |
| "learning_rate": 0.0019862007473534027, | |
| "loss": 0.3028, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.1947431302270013, | |
| "grad_norm": 0.10546875, | |
| "learning_rate": 0.001982973099683902, | |
| "loss": 0.3006, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2544802867383513, | |
| "grad_norm": 0.1044921875, | |
| "learning_rate": 0.001979409767601366, | |
| "loss": 0.2981, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.3142174432497014, | |
| "grad_norm": 0.10888671875, | |
| "learning_rate": 0.001975511967980437, | |
| "loss": 0.2917, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.3739545997610514, | |
| "grad_norm": 0.10107421875, | |
| "learning_rate": 0.001971281031916114, | |
| "loss": 0.2816, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.4336917562724014, | |
| "grad_norm": 0.1337890625, | |
| "learning_rate": 0.0019667184042691877, | |
| "loss": 0.2754, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.4934289127837514, | |
| "grad_norm": 0.08203125, | |
| "learning_rate": 0.001961825643172819, | |
| "loss": 0.2775, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.5531660692951015, | |
| "grad_norm": 0.0634765625, | |
| "learning_rate": 0.0019566044195004407, | |
| "loss": 0.2756, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 0.1142578125, | |
| "learning_rate": 0.0019510565162951536, | |
| "loss": 0.27, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.6726403823178018, | |
| "grad_norm": 0.09423828125, | |
| "learning_rate": 0.0019451838281608197, | |
| "loss": 0.2616, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.7323775388291518, | |
| "grad_norm": 0.095703125, | |
| "learning_rate": 0.0019389883606150567, | |
| "loss": 0.2589, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.7921146953405018, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.0019324722294043557, | |
| "loss": 0.2567, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.8518518518518519, | |
| "grad_norm": 0.138671875, | |
| "learning_rate": 0.0019256376597815564, | |
| "loss": 0.2462, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.911589008363202, | |
| "grad_norm": 0.10205078125, | |
| "learning_rate": 0.001918486985745923, | |
| "loss": 0.246, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.971326164874552, | |
| "grad_norm": 0.07861328125, | |
| "learning_rate": 0.0019110226492460884, | |
| "loss": 0.2378, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.031063321385902, | |
| "grad_norm": 0.091796875, | |
| "learning_rate": 0.0019032471993461289, | |
| "loss": 0.2322, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.090800477897252, | |
| "grad_norm": 0.09814453125, | |
| "learning_rate": 0.0018951632913550625, | |
| "loss": 0.2341, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.150537634408602, | |
| "grad_norm": 0.0859375, | |
| "learning_rate": 0.0018867736859200619, | |
| "loss": 0.2314, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.2102747909199523, | |
| "grad_norm": 0.09765625, | |
| "learning_rate": 0.0018780812480836979, | |
| "loss": 0.2297, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.270011947431302, | |
| "grad_norm": 0.087890625, | |
| "learning_rate": 0.0018690889463055284, | |
| "loss": 0.224, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.3297491039426523, | |
| "grad_norm": 0.0693359375, | |
| "learning_rate": 0.0018597998514483724, | |
| "loss": 0.2193, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.3894862604540026, | |
| "grad_norm": 0.11669921875, | |
| "learning_rate": 0.0018502171357296143, | |
| "loss": 0.2158, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.4492234169653524, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 0.0018403440716378927, | |
| "loss": 0.2081, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.5089605734767026, | |
| "grad_norm": 0.10498046875, | |
| "learning_rate": 0.0018301840308155505, | |
| "loss": 0.2092, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.5686977299880525, | |
| "grad_norm": 0.10107421875, | |
| "learning_rate": 0.0018197404829072212, | |
| "loss": 0.2038, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.6284348864994027, | |
| "grad_norm": 0.103515625, | |
| "learning_rate": 0.0018090169943749475, | |
| "loss": 0.199, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.6881720430107525, | |
| "grad_norm": 0.087890625, | |
| "learning_rate": 0.0017980172272802398, | |
| "loss": 0.1955, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.7479091995221028, | |
| "grad_norm": 0.0986328125, | |
| "learning_rate": 0.0017867449380334832, | |
| "loss": 0.1937, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.807646356033453, | |
| "grad_norm": 0.1044921875, | |
| "learning_rate": 0.0017752039761111298, | |
| "loss": 0.1914, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.867383512544803, | |
| "grad_norm": 0.0947265625, | |
| "learning_rate": 0.001763398282741103, | |
| "loss": 0.1875, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.927120669056153, | |
| "grad_norm": 0.09765625, | |
| "learning_rate": 0.0017513318895568735, | |
| "loss": 0.1855, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.986857825567503, | |
| "grad_norm": 0.08056640625, | |
| "learning_rate": 0.001739008917220659, | |
| "loss": 0.1795, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.046594982078853, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 0.0017264335740162242, | |
| "loss": 0.1742, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.106332138590203, | |
| "grad_norm": 0.09228515625, | |
| "learning_rate": 0.0017136101544117524, | |
| "loss": 0.1759, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.166069295101553, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 0.0017005430375932908, | |
| "loss": 0.1756, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "grad_norm": 0.07470703125, | |
| "learning_rate": 0.0016872366859692627, | |
| "loss": 0.1766, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.2855436081242533, | |
| "grad_norm": 0.087890625, | |
| "learning_rate": 0.0016736956436465573, | |
| "loss": 0.1666, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.3452807646356035, | |
| "grad_norm": 0.0869140625, | |
| "learning_rate": 0.0016599245348787228, | |
| "loss": 0.169, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.4050179211469533, | |
| "grad_norm": 0.08740234375, | |
| "learning_rate": 0.0016459280624867873, | |
| "loss": 0.1616, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.4647550776583036, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 0.001631711006253251, | |
| "loss": 0.1595, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.5244922341696534, | |
| "grad_norm": 0.078125, | |
| "learning_rate": 0.001617278221289793, | |
| "loss": 0.1626, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.5842293906810037, | |
| "grad_norm": 0.0908203125, | |
| "learning_rate": 0.0016026346363792565, | |
| "loss": 0.1546, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 18000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": true, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1559951680731546e+17, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |