lesso07's picture
Training in progress, step 75, checkpoint
39029de verified
raw
history blame
15.6 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.01686719892049927,
"eval_steps": 9,
"global_step": 75,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0002248959856066569,
"grad_norm": 0.641430139541626,
"learning_rate": 1e-05,
"loss": 0.2596,
"step": 1
},
{
"epoch": 0.0002248959856066569,
"eval_loss": 0.24757881462574005,
"eval_runtime": 820.997,
"eval_samples_per_second": 4.562,
"eval_steps_per_second": 0.571,
"step": 1
},
{
"epoch": 0.0004497919712133138,
"grad_norm": 0.38873377442359924,
"learning_rate": 2e-05,
"loss": 0.2105,
"step": 2
},
{
"epoch": 0.0006746879568199708,
"grad_norm": 0.5534820556640625,
"learning_rate": 3e-05,
"loss": 0.2969,
"step": 3
},
{
"epoch": 0.0008995839424266276,
"grad_norm": 0.4102833867073059,
"learning_rate": 4e-05,
"loss": 0.2417,
"step": 4
},
{
"epoch": 0.0011244799280332846,
"grad_norm": 0.4007568657398224,
"learning_rate": 5e-05,
"loss": 0.2487,
"step": 5
},
{
"epoch": 0.0013493759136399416,
"grad_norm": 0.46972644329071045,
"learning_rate": 6e-05,
"loss": 0.2449,
"step": 6
},
{
"epoch": 0.0015742718992465985,
"grad_norm": 0.5923429727554321,
"learning_rate": 7e-05,
"loss": 0.2189,
"step": 7
},
{
"epoch": 0.0017991678848532553,
"grad_norm": 0.41178950667381287,
"learning_rate": 8e-05,
"loss": 0.1625,
"step": 8
},
{
"epoch": 0.0020240638704599125,
"grad_norm": 0.3475341498851776,
"learning_rate": 9e-05,
"loss": 0.137,
"step": 9
},
{
"epoch": 0.0020240638704599125,
"eval_loss": 0.1294211745262146,
"eval_runtime": 821.6488,
"eval_samples_per_second": 4.558,
"eval_steps_per_second": 0.571,
"step": 9
},
{
"epoch": 0.0022489598560665692,
"grad_norm": 0.25473251938819885,
"learning_rate": 0.0001,
"loss": 0.1002,
"step": 10
},
{
"epoch": 0.002473855841673226,
"grad_norm": 0.27986106276512146,
"learning_rate": 9.99695413509548e-05,
"loss": 0.0978,
"step": 11
},
{
"epoch": 0.002698751827279883,
"grad_norm": 0.33775243163108826,
"learning_rate": 9.987820251299122e-05,
"loss": 0.1088,
"step": 12
},
{
"epoch": 0.00292364781288654,
"grad_norm": 0.41962045431137085,
"learning_rate": 9.972609476841367e-05,
"loss": 0.0992,
"step": 13
},
{
"epoch": 0.003148543798493197,
"grad_norm": 0.2988746762275696,
"learning_rate": 9.951340343707852e-05,
"loss": 0.0641,
"step": 14
},
{
"epoch": 0.003373439784099854,
"grad_norm": 0.4276213049888611,
"learning_rate": 9.924038765061042e-05,
"loss": 0.0991,
"step": 15
},
{
"epoch": 0.0035983357697065106,
"grad_norm": 0.31324365735054016,
"learning_rate": 9.890738003669029e-05,
"loss": 0.0529,
"step": 16
},
{
"epoch": 0.0038232317553131678,
"grad_norm": 0.49720054864883423,
"learning_rate": 9.851478631379982e-05,
"loss": 0.0674,
"step": 17
},
{
"epoch": 0.004048127740919825,
"grad_norm": 0.4947662353515625,
"learning_rate": 9.806308479691595e-05,
"loss": 0.0473,
"step": 18
},
{
"epoch": 0.004048127740919825,
"eval_loss": 0.044178664684295654,
"eval_runtime": 821.2855,
"eval_samples_per_second": 4.56,
"eval_steps_per_second": 0.571,
"step": 18
},
{
"epoch": 0.004273023726526481,
"grad_norm": 0.31270837783813477,
"learning_rate": 9.755282581475769e-05,
"loss": 0.0364,
"step": 19
},
{
"epoch": 0.0044979197121331384,
"grad_norm": 0.3267187476158142,
"learning_rate": 9.698463103929542e-05,
"loss": 0.0485,
"step": 20
},
{
"epoch": 0.004722815697739796,
"grad_norm": 0.24129141867160797,
"learning_rate": 9.635919272833938e-05,
"loss": 0.0372,
"step": 21
},
{
"epoch": 0.004947711683346452,
"grad_norm": 0.29355746507644653,
"learning_rate": 9.567727288213005e-05,
"loss": 0.0342,
"step": 22
},
{
"epoch": 0.005172607668953109,
"grad_norm": 0.4001331627368927,
"learning_rate": 9.493970231495835e-05,
"loss": 0.0246,
"step": 23
},
{
"epoch": 0.005397503654559766,
"grad_norm": 0.22919444739818573,
"learning_rate": 9.414737964294636e-05,
"loss": 0.0263,
"step": 24
},
{
"epoch": 0.005622399640166423,
"grad_norm": 0.20601242780685425,
"learning_rate": 9.330127018922194e-05,
"loss": 0.0197,
"step": 25
},
{
"epoch": 0.00584729562577308,
"grad_norm": 0.18810567259788513,
"learning_rate": 9.24024048078213e-05,
"loss": 0.0156,
"step": 26
},
{
"epoch": 0.006072191611379737,
"grad_norm": 0.3696419298648834,
"learning_rate": 9.145187862775209e-05,
"loss": 0.0168,
"step": 27
},
{
"epoch": 0.006072191611379737,
"eval_loss": 0.01646805927157402,
"eval_runtime": 821.6288,
"eval_samples_per_second": 4.558,
"eval_steps_per_second": 0.571,
"step": 27
},
{
"epoch": 0.006297087596986394,
"grad_norm": 0.25218066573143005,
"learning_rate": 9.045084971874738e-05,
"loss": 0.0093,
"step": 28
},
{
"epoch": 0.0065219835825930505,
"grad_norm": 0.34651580452919006,
"learning_rate": 8.940053768033609e-05,
"loss": 0.0176,
"step": 29
},
{
"epoch": 0.006746879568199708,
"grad_norm": 0.36292922496795654,
"learning_rate": 8.83022221559489e-05,
"loss": 0.0124,
"step": 30
},
{
"epoch": 0.006971775553806365,
"grad_norm": 0.19089706242084503,
"learning_rate": 8.715724127386972e-05,
"loss": 0.0203,
"step": 31
},
{
"epoch": 0.007196671539413021,
"grad_norm": 0.24353894591331482,
"learning_rate": 8.596699001693255e-05,
"loss": 0.0101,
"step": 32
},
{
"epoch": 0.007421567525019678,
"grad_norm": 0.23627233505249023,
"learning_rate": 8.473291852294987e-05,
"loss": 0.0179,
"step": 33
},
{
"epoch": 0.0076464635106263355,
"grad_norm": 0.4310121238231659,
"learning_rate": 8.345653031794292e-05,
"loss": 0.0214,
"step": 34
},
{
"epoch": 0.007871359496232992,
"grad_norm": 0.227275088429451,
"learning_rate": 8.213938048432697e-05,
"loss": 0.0098,
"step": 35
},
{
"epoch": 0.00809625548183965,
"grad_norm": 0.2399313896894455,
"learning_rate": 8.07830737662829e-05,
"loss": 0.0111,
"step": 36
},
{
"epoch": 0.00809625548183965,
"eval_loss": 0.011674784123897552,
"eval_runtime": 821.5999,
"eval_samples_per_second": 4.558,
"eval_steps_per_second": 0.571,
"step": 36
},
{
"epoch": 0.008321151467446306,
"grad_norm": 0.1582072526216507,
"learning_rate": 7.938926261462366e-05,
"loss": 0.0038,
"step": 37
},
{
"epoch": 0.008546047453052963,
"grad_norm": 0.24019911885261536,
"learning_rate": 7.795964517353735e-05,
"loss": 0.0069,
"step": 38
},
{
"epoch": 0.00877094343865962,
"grad_norm": 0.20569612085819244,
"learning_rate": 7.649596321166024e-05,
"loss": 0.0065,
"step": 39
},
{
"epoch": 0.008995839424266277,
"grad_norm": 0.30835479497909546,
"learning_rate": 7.500000000000001e-05,
"loss": 0.0073,
"step": 40
},
{
"epoch": 0.009220735409872933,
"grad_norm": 0.25597208738327026,
"learning_rate": 7.347357813929454e-05,
"loss": 0.0127,
"step": 41
},
{
"epoch": 0.009445631395479591,
"grad_norm": 0.09658183157444,
"learning_rate": 7.191855733945387e-05,
"loss": 0.0047,
"step": 42
},
{
"epoch": 0.009670527381086248,
"grad_norm": 0.20138397812843323,
"learning_rate": 7.033683215379002e-05,
"loss": 0.0162,
"step": 43
},
{
"epoch": 0.009895423366692904,
"grad_norm": 0.2869853377342224,
"learning_rate": 6.873032967079561e-05,
"loss": 0.0074,
"step": 44
},
{
"epoch": 0.010120319352299562,
"grad_norm": 0.3238297700881958,
"learning_rate": 6.710100716628344e-05,
"loss": 0.022,
"step": 45
},
{
"epoch": 0.010120319352299562,
"eval_loss": 0.012101165018975735,
"eval_runtime": 821.4174,
"eval_samples_per_second": 4.559,
"eval_steps_per_second": 0.571,
"step": 45
},
{
"epoch": 0.010345215337906218,
"grad_norm": 0.1724397987127304,
"learning_rate": 6.545084971874738e-05,
"loss": 0.0064,
"step": 46
},
{
"epoch": 0.010570111323512875,
"grad_norm": 0.3781624138355255,
"learning_rate": 6.378186779084995e-05,
"loss": 0.0113,
"step": 47
},
{
"epoch": 0.010795007309119533,
"grad_norm": 0.17849580943584442,
"learning_rate": 6.209609477998338e-05,
"loss": 0.0067,
"step": 48
},
{
"epoch": 0.011019903294726189,
"grad_norm": 0.19034115970134735,
"learning_rate": 6.0395584540887963e-05,
"loss": 0.006,
"step": 49
},
{
"epoch": 0.011244799280332845,
"grad_norm": 0.5638530254364014,
"learning_rate": 5.868240888334653e-05,
"loss": 0.019,
"step": 50
},
{
"epoch": 0.011469695265939503,
"grad_norm": 0.1782771646976471,
"learning_rate": 5.695865504800327e-05,
"loss": 0.005,
"step": 51
},
{
"epoch": 0.01169459125154616,
"grad_norm": 0.19799622893333435,
"learning_rate": 5.522642316338268e-05,
"loss": 0.0081,
"step": 52
},
{
"epoch": 0.011919487237152818,
"grad_norm": 0.3330150544643402,
"learning_rate": 5.348782368720626e-05,
"loss": 0.0126,
"step": 53
},
{
"epoch": 0.012144383222759474,
"grad_norm": 0.5607080459594727,
"learning_rate": 5.174497483512506e-05,
"loss": 0.0103,
"step": 54
},
{
"epoch": 0.012144383222759474,
"eval_loss": 0.009038399904966354,
"eval_runtime": 821.8702,
"eval_samples_per_second": 4.557,
"eval_steps_per_second": 0.571,
"step": 54
},
{
"epoch": 0.01236927920836613,
"grad_norm": 0.3719494938850403,
"learning_rate": 5e-05,
"loss": 0.0095,
"step": 55
},
{
"epoch": 0.012594175193972788,
"grad_norm": 0.29955190420150757,
"learning_rate": 4.825502516487497e-05,
"loss": 0.0129,
"step": 56
},
{
"epoch": 0.012819071179579445,
"grad_norm": 0.4373495876789093,
"learning_rate": 4.6512176312793736e-05,
"loss": 0.0102,
"step": 57
},
{
"epoch": 0.013043967165186101,
"grad_norm": 0.20729227364063263,
"learning_rate": 4.477357683661734e-05,
"loss": 0.0096,
"step": 58
},
{
"epoch": 0.013268863150792759,
"grad_norm": 0.22739438712596893,
"learning_rate": 4.3041344951996746e-05,
"loss": 0.0041,
"step": 59
},
{
"epoch": 0.013493759136399415,
"grad_norm": 0.42493709921836853,
"learning_rate": 4.131759111665349e-05,
"loss": 0.0085,
"step": 60
},
{
"epoch": 0.013718655122006072,
"grad_norm": 0.23612692952156067,
"learning_rate": 3.960441545911204e-05,
"loss": 0.0095,
"step": 61
},
{
"epoch": 0.01394355110761273,
"grad_norm": 0.3186267912387848,
"learning_rate": 3.790390522001662e-05,
"loss": 0.015,
"step": 62
},
{
"epoch": 0.014168447093219386,
"grad_norm": 0.33325353264808655,
"learning_rate": 3.6218132209150045e-05,
"loss": 0.0051,
"step": 63
},
{
"epoch": 0.014168447093219386,
"eval_loss": 0.007068150211125612,
"eval_runtime": 822.03,
"eval_samples_per_second": 4.556,
"eval_steps_per_second": 0.571,
"step": 63
},
{
"epoch": 0.014393343078826042,
"grad_norm": 0.34111836552619934,
"learning_rate": 3.4549150281252636e-05,
"loss": 0.0217,
"step": 64
},
{
"epoch": 0.0146182390644327,
"grad_norm": 0.3350174129009247,
"learning_rate": 3.289899283371657e-05,
"loss": 0.0088,
"step": 65
},
{
"epoch": 0.014843135050039357,
"grad_norm": 0.11381199955940247,
"learning_rate": 3.12696703292044e-05,
"loss": 0.0062,
"step": 66
},
{
"epoch": 0.015068031035646013,
"grad_norm": 0.19116997718811035,
"learning_rate": 2.9663167846209998e-05,
"loss": 0.0029,
"step": 67
},
{
"epoch": 0.015292927021252671,
"grad_norm": 0.2507087290287018,
"learning_rate": 2.8081442660546125e-05,
"loss": 0.0129,
"step": 68
},
{
"epoch": 0.015517823006859327,
"grad_norm": 0.21788032352924347,
"learning_rate": 2.6526421860705473e-05,
"loss": 0.005,
"step": 69
},
{
"epoch": 0.015742718992465984,
"grad_norm": 0.15248437225818634,
"learning_rate": 2.500000000000001e-05,
"loss": 0.0019,
"step": 70
},
{
"epoch": 0.015967614978072642,
"grad_norm": 0.4119681417942047,
"learning_rate": 2.350403678833976e-05,
"loss": 0.0078,
"step": 71
},
{
"epoch": 0.0161925109636793,
"grad_norm": 0.13133472204208374,
"learning_rate": 2.2040354826462668e-05,
"loss": 0.0067,
"step": 72
},
{
"epoch": 0.0161925109636793,
"eval_loss": 0.005652678664773703,
"eval_runtime": 821.9265,
"eval_samples_per_second": 4.556,
"eval_steps_per_second": 0.571,
"step": 72
},
{
"epoch": 0.016417406949285954,
"grad_norm": 0.11950453370809555,
"learning_rate": 2.061073738537635e-05,
"loss": 0.0028,
"step": 73
},
{
"epoch": 0.016642302934892612,
"grad_norm": 0.15952032804489136,
"learning_rate": 1.9216926233717085e-05,
"loss": 0.0055,
"step": 74
},
{
"epoch": 0.01686719892049927,
"grad_norm": 0.28256332874298096,
"learning_rate": 1.7860619515673033e-05,
"loss": 0.002,
"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": 1.097993179669463e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}