dzanbek's picture
Training in progress, step 50, checkpoint
b274be1 verified
raw
history blame
9.47 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.001239552893271397,
"eval_steps": 500,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 2.479105786542794e-05,
"grad_norm": 7.298285007476807,
"learning_rate": 0.0001,
"loss": 8.3458,
"step": 1
},
{
"epoch": 4.958211573085588e-05,
"grad_norm": 8.879791259765625,
"learning_rate": 0.0002,
"loss": 8.3615,
"step": 2
},
{
"epoch": 7.437317359628382e-05,
"grad_norm": 8.794269561767578,
"learning_rate": 0.00019978589232386035,
"loss": 7.5151,
"step": 3
},
{
"epoch": 9.916423146171176e-05,
"grad_norm": 9.012420654296875,
"learning_rate": 0.00019914448613738106,
"loss": 5.6806,
"step": 4
},
{
"epoch": 0.0001239552893271397,
"grad_norm": 7.061407566070557,
"learning_rate": 0.00019807852804032305,
"loss": 3.9168,
"step": 5
},
{
"epoch": 0.00014874634719256764,
"grad_norm": 6.04531717300415,
"learning_rate": 0.00019659258262890683,
"loss": 2.518,
"step": 6
},
{
"epoch": 0.00017353740505799558,
"grad_norm": 5.078192234039307,
"learning_rate": 0.0001946930129495106,
"loss": 1.6634,
"step": 7
},
{
"epoch": 0.00019832846292342353,
"grad_norm": 5.033928871154785,
"learning_rate": 0.0001923879532511287,
"loss": 0.907,
"step": 8
},
{
"epoch": 0.00022311952078885147,
"grad_norm": 3.0206494331359863,
"learning_rate": 0.00018968727415326884,
"loss": 0.6279,
"step": 9
},
{
"epoch": 0.0002479105786542794,
"grad_norm": 2.810877799987793,
"learning_rate": 0.00018660254037844388,
"loss": 0.419,
"step": 10
},
{
"epoch": 0.00027270163651970736,
"grad_norm": 3.125711441040039,
"learning_rate": 0.00018314696123025454,
"loss": 0.7372,
"step": 11
},
{
"epoch": 0.0002974926943851353,
"grad_norm": 3.9015979766845703,
"learning_rate": 0.00017933533402912354,
"loss": 0.312,
"step": 12
},
{
"epoch": 0.0003222837522505632,
"grad_norm": 1.802170753479004,
"learning_rate": 0.00017518398074789775,
"loss": 0.1449,
"step": 13
},
{
"epoch": 0.00034707481011599117,
"grad_norm": 1.4817924499511719,
"learning_rate": 0.00017071067811865476,
"loss": 0.0737,
"step": 14
},
{
"epoch": 0.0003718658679814191,
"grad_norm": 0.31127655506134033,
"learning_rate": 0.00016593458151000688,
"loss": 0.0081,
"step": 15
},
{
"epoch": 0.00039665692584684706,
"grad_norm": 1.7345539331436157,
"learning_rate": 0.00016087614290087208,
"loss": 0.1763,
"step": 16
},
{
"epoch": 0.000421447983712275,
"grad_norm": 0.861487865447998,
"learning_rate": 0.00015555702330196023,
"loss": 0.0059,
"step": 17
},
{
"epoch": 0.00044623904157770294,
"grad_norm": 0.014484393410384655,
"learning_rate": 0.00015000000000000001,
"loss": 0.0003,
"step": 18
},
{
"epoch": 0.00047103009944313086,
"grad_norm": 0.020801691338419914,
"learning_rate": 0.00014422886902190014,
"loss": 0.0003,
"step": 19
},
{
"epoch": 0.0004958211573085588,
"grad_norm": 0.035404324531555176,
"learning_rate": 0.000138268343236509,
"loss": 0.0003,
"step": 20
},
{
"epoch": 0.0005206122151739868,
"grad_norm": 15.371546745300293,
"learning_rate": 0.00013214394653031616,
"loss": 0.1066,
"step": 21
},
{
"epoch": 0.0005454032730394147,
"grad_norm": 0.01011655479669571,
"learning_rate": 0.00012588190451025207,
"loss": 0.0002,
"step": 22
},
{
"epoch": 0.0005701943309048426,
"grad_norm": 0.09178952872753143,
"learning_rate": 0.00011950903220161285,
"loss": 0.001,
"step": 23
},
{
"epoch": 0.0005949853887702706,
"grad_norm": 0.00701731164008379,
"learning_rate": 0.00011305261922200519,
"loss": 0.0001,
"step": 24
},
{
"epoch": 0.0006197764466356985,
"grad_norm": 0.014851056970655918,
"learning_rate": 0.00010654031292301432,
"loss": 0.0003,
"step": 25
},
{
"epoch": 0.0006445675045011264,
"grad_norm": 2.1300110816955566,
"learning_rate": 0.0001,
"loss": 0.0058,
"step": 26
},
{
"epoch": 0.0006693585623665544,
"grad_norm": 1.1658985614776611,
"learning_rate": 9.345968707698569e-05,
"loss": 0.0137,
"step": 27
},
{
"epoch": 0.0006941496202319823,
"grad_norm": 0.008638538420200348,
"learning_rate": 8.694738077799488e-05,
"loss": 0.0001,
"step": 28
},
{
"epoch": 0.0007189406780974103,
"grad_norm": 0.0034529389813542366,
"learning_rate": 8.049096779838719e-05,
"loss": 0.0001,
"step": 29
},
{
"epoch": 0.0007437317359628382,
"grad_norm": 0.0020151117350906134,
"learning_rate": 7.411809548974792e-05,
"loss": 0.0,
"step": 30
},
{
"epoch": 0.0007685227938282661,
"grad_norm": 2.9357662200927734,
"learning_rate": 6.785605346968386e-05,
"loss": 0.1847,
"step": 31
},
{
"epoch": 0.0007933138516936941,
"grad_norm": 1.7568498849868774,
"learning_rate": 6.173165676349103e-05,
"loss": 0.0447,
"step": 32
},
{
"epoch": 0.000818104909559122,
"grad_norm": 1.2945858240127563,
"learning_rate": 5.577113097809989e-05,
"loss": 0.029,
"step": 33
},
{
"epoch": 0.00084289596742455,
"grad_norm": 0.003499679733067751,
"learning_rate": 5.000000000000002e-05,
"loss": 0.0001,
"step": 34
},
{
"epoch": 0.0008676870252899779,
"grad_norm": 0.0012576651060953736,
"learning_rate": 4.444297669803981e-05,
"loss": 0.0,
"step": 35
},
{
"epoch": 0.0008924780831554059,
"grad_norm": 0.0027500190772116184,
"learning_rate": 3.9123857099127936e-05,
"loss": 0.0001,
"step": 36
},
{
"epoch": 0.0009172691410208338,
"grad_norm": 0.004673486575484276,
"learning_rate": 3.406541848999312e-05,
"loss": 0.0001,
"step": 37
},
{
"epoch": 0.0009420601988862617,
"grad_norm": 0.0062264916487038136,
"learning_rate": 2.9289321881345254e-05,
"loss": 0.0001,
"step": 38
},
{
"epoch": 0.0009668512567516897,
"grad_norm": 0.001728757400996983,
"learning_rate": 2.4816019252102273e-05,
"loss": 0.0,
"step": 39
},
{
"epoch": 0.0009916423146171176,
"grad_norm": 0.004961659200489521,
"learning_rate": 2.0664665970876496e-05,
"loss": 0.0001,
"step": 40
},
{
"epoch": 0.0010164333724825456,
"grad_norm": 0.0015779470559209585,
"learning_rate": 1.6853038769745467e-05,
"loss": 0.0,
"step": 41
},
{
"epoch": 0.0010412244303479735,
"grad_norm": 0.0018025401514023542,
"learning_rate": 1.339745962155613e-05,
"loss": 0.0,
"step": 42
},
{
"epoch": 0.0010660154882134014,
"grad_norm": 0.0029116978403180838,
"learning_rate": 1.0312725846731175e-05,
"loss": 0.0,
"step": 43
},
{
"epoch": 0.0010908065460788294,
"grad_norm": 0.0026793968863785267,
"learning_rate": 7.612046748871327e-06,
"loss": 0.0,
"step": 44
},
{
"epoch": 0.0011155976039442573,
"grad_norm": 0.002206047298386693,
"learning_rate": 5.306987050489442e-06,
"loss": 0.0,
"step": 45
},
{
"epoch": 0.0011403886618096852,
"grad_norm": 0.002513278741389513,
"learning_rate": 3.40741737109318e-06,
"loss": 0.0,
"step": 46
},
{
"epoch": 0.0011651797196751133,
"grad_norm": 0.00416777515783906,
"learning_rate": 1.921471959676957e-06,
"loss": 0.0001,
"step": 47
},
{
"epoch": 0.0011899707775405411,
"grad_norm": 0.012963655404746532,
"learning_rate": 8.555138626189618e-07,
"loss": 0.0002,
"step": 48
},
{
"epoch": 0.001214761835405969,
"grad_norm": 0.001530441571958363,
"learning_rate": 2.141076761396521e-07,
"loss": 0.0,
"step": 49
},
{
"epoch": 0.001239552893271397,
"grad_norm": 0.002721717581152916,
"learning_rate": 0.0,
"loss": 0.0001,
"step": 50
}
],
"logging_steps": 1,
"max_steps": 50,
"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": 6.6109411688448e+16,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}