masatochi's picture
Training in progress, step 90, checkpoint
012d4b2 verified
raw
history blame
16.9 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.04401247019988997,
"eval_steps": 34,
"global_step": 90,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0004890274466654441,
"grad_norm": 0.42223185300827026,
"learning_rate": 6.666666666666667e-06,
"loss": 1.2369,
"step": 1
},
{
"epoch": 0.0004890274466654441,
"eval_loss": 1.3036277294158936,
"eval_runtime": 1355.2246,
"eval_samples_per_second": 1.906,
"eval_steps_per_second": 0.635,
"step": 1
},
{
"epoch": 0.0009780548933308881,
"grad_norm": 0.49382174015045166,
"learning_rate": 1.3333333333333333e-05,
"loss": 1.3544,
"step": 2
},
{
"epoch": 0.0014670823399963323,
"grad_norm": 0.43727555871009827,
"learning_rate": 2e-05,
"loss": 1.403,
"step": 3
},
{
"epoch": 0.0019561097866617762,
"grad_norm": 0.3653078079223633,
"learning_rate": 2.6666666666666667e-05,
"loss": 1.189,
"step": 4
},
{
"epoch": 0.0024451372333272204,
"grad_norm": 0.3689245283603668,
"learning_rate": 3.3333333333333335e-05,
"loss": 1.1686,
"step": 5
},
{
"epoch": 0.0029341646799926646,
"grad_norm": 0.44767728447914124,
"learning_rate": 4e-05,
"loss": 1.2578,
"step": 6
},
{
"epoch": 0.0034231921266581087,
"grad_norm": 0.4529785215854645,
"learning_rate": 4.666666666666667e-05,
"loss": 1.2452,
"step": 7
},
{
"epoch": 0.0039122195733235525,
"grad_norm": 0.6146576404571533,
"learning_rate": 5.333333333333333e-05,
"loss": 1.5423,
"step": 8
},
{
"epoch": 0.004401247019988997,
"grad_norm": 0.42456191778182983,
"learning_rate": 6e-05,
"loss": 1.1323,
"step": 9
},
{
"epoch": 0.004890274466654441,
"grad_norm": 0.4658707082271576,
"learning_rate": 6.666666666666667e-05,
"loss": 1.2595,
"step": 10
},
{
"epoch": 0.005379301913319885,
"grad_norm": 0.6462928652763367,
"learning_rate": 7.333333333333333e-05,
"loss": 1.3136,
"step": 11
},
{
"epoch": 0.005868329359985329,
"grad_norm": 0.514143168926239,
"learning_rate": 8e-05,
"loss": 1.1999,
"step": 12
},
{
"epoch": 0.006357356806650773,
"grad_norm": 0.43792784214019775,
"learning_rate": 8.666666666666667e-05,
"loss": 1.0675,
"step": 13
},
{
"epoch": 0.0068463842533162175,
"grad_norm": 0.6212848424911499,
"learning_rate": 9.333333333333334e-05,
"loss": 1.3066,
"step": 14
},
{
"epoch": 0.007335411699981661,
"grad_norm": 0.6186118125915527,
"learning_rate": 0.0001,
"loss": 1.2355,
"step": 15
},
{
"epoch": 0.007824439146647105,
"grad_norm": 0.6825079917907715,
"learning_rate": 0.00010666666666666667,
"loss": 1.2258,
"step": 16
},
{
"epoch": 0.00831346659331255,
"grad_norm": 0.3936300575733185,
"learning_rate": 0.00011333333333333334,
"loss": 1.1337,
"step": 17
},
{
"epoch": 0.008802494039977994,
"grad_norm": 0.5038918852806091,
"learning_rate": 0.00012,
"loss": 1.1278,
"step": 18
},
{
"epoch": 0.009291521486643438,
"grad_norm": 0.5894134640693665,
"learning_rate": 0.00012666666666666666,
"loss": 1.1902,
"step": 19
},
{
"epoch": 0.009780548933308882,
"grad_norm": 1.1452070474624634,
"learning_rate": 0.00013333333333333334,
"loss": 1.1023,
"step": 20
},
{
"epoch": 0.010269576379974325,
"grad_norm": 0.4289817214012146,
"learning_rate": 0.00014,
"loss": 1.0195,
"step": 21
},
{
"epoch": 0.01075860382663977,
"grad_norm": 1.290226936340332,
"learning_rate": 0.00014666666666666666,
"loss": 0.9534,
"step": 22
},
{
"epoch": 0.011247631273305215,
"grad_norm": 0.5022335648536682,
"learning_rate": 0.00015333333333333334,
"loss": 0.8952,
"step": 23
},
{
"epoch": 0.011736658719970658,
"grad_norm": 0.5421992540359497,
"learning_rate": 0.00016,
"loss": 0.8553,
"step": 24
},
{
"epoch": 0.012225686166636102,
"grad_norm": 1.568458080291748,
"learning_rate": 0.0001666666666666667,
"loss": 0.9764,
"step": 25
},
{
"epoch": 0.012714713613301546,
"grad_norm": 0.6190416812896729,
"learning_rate": 0.00017333333333333334,
"loss": 0.9819,
"step": 26
},
{
"epoch": 0.013203741059966991,
"grad_norm": 0.4747830927371979,
"learning_rate": 0.00018,
"loss": 0.9395,
"step": 27
},
{
"epoch": 0.013692768506632435,
"grad_norm": 0.9486120343208313,
"learning_rate": 0.0001866666666666667,
"loss": 0.9669,
"step": 28
},
{
"epoch": 0.014181795953297879,
"grad_norm": 0.4615766108036041,
"learning_rate": 0.00019333333333333333,
"loss": 0.9341,
"step": 29
},
{
"epoch": 0.014670823399963322,
"grad_norm": 0.46146905422210693,
"learning_rate": 0.0002,
"loss": 1.0082,
"step": 30
},
{
"epoch": 0.015159850846628768,
"grad_norm": 0.47625860571861267,
"learning_rate": 0.00019998292504580528,
"loss": 0.9419,
"step": 31
},
{
"epoch": 0.01564887829329421,
"grad_norm": 0.5290003418922424,
"learning_rate": 0.0001999317060143023,
"loss": 1.1635,
"step": 32
},
{
"epoch": 0.016137905739959654,
"grad_norm": 0.5592769980430603,
"learning_rate": 0.0001998463603967434,
"loss": 0.9437,
"step": 33
},
{
"epoch": 0.0166269331866251,
"grad_norm": 0.3583241403102875,
"learning_rate": 0.00019972691733857883,
"loss": 1.1428,
"step": 34
},
{
"epoch": 0.0166269331866251,
"eval_loss": 1.0231887102127075,
"eval_runtime": 1359.3297,
"eval_samples_per_second": 1.9,
"eval_steps_per_second": 0.633,
"step": 34
},
{
"epoch": 0.017115960633290545,
"grad_norm": 0.41851457953453064,
"learning_rate": 0.00019957341762950344,
"loss": 1.0976,
"step": 35
},
{
"epoch": 0.017604988079955988,
"grad_norm": 0.46739837527275085,
"learning_rate": 0.0001993859136895274,
"loss": 1.0089,
"step": 36
},
{
"epoch": 0.018094015526621432,
"grad_norm": 0.3970998227596283,
"learning_rate": 0.00019916446955107428,
"loss": 1.1005,
"step": 37
},
{
"epoch": 0.018583042973286876,
"grad_norm": 0.5996494293212891,
"learning_rate": 0.0001989091608371146,
"loss": 0.9741,
"step": 38
},
{
"epoch": 0.01907207041995232,
"grad_norm": 0.27929016947746277,
"learning_rate": 0.00019862007473534025,
"loss": 0.9607,
"step": 39
},
{
"epoch": 0.019561097866617763,
"grad_norm": 0.33169642090797424,
"learning_rate": 0.0001982973099683902,
"loss": 1.1173,
"step": 40
},
{
"epoch": 0.020050125313283207,
"grad_norm": 0.3042043149471283,
"learning_rate": 0.0001979409767601366,
"loss": 1.1178,
"step": 41
},
{
"epoch": 0.02053915275994865,
"grad_norm": 0.3394615650177002,
"learning_rate": 0.00019755119679804367,
"loss": 1.0167,
"step": 42
},
{
"epoch": 0.021028180206614098,
"grad_norm": 0.3241017758846283,
"learning_rate": 0.0001971281031916114,
"loss": 1.0089,
"step": 43
},
{
"epoch": 0.02151720765327954,
"grad_norm": 0.6870384812355042,
"learning_rate": 0.00019667184042691875,
"loss": 1.0362,
"step": 44
},
{
"epoch": 0.022006235099944985,
"grad_norm": 0.31295979022979736,
"learning_rate": 0.00019618256431728194,
"loss": 1.0032,
"step": 45
},
{
"epoch": 0.02249526254661043,
"grad_norm": 0.32893508672714233,
"learning_rate": 0.0001956604419500441,
"loss": 1.0651,
"step": 46
},
{
"epoch": 0.022984289993275873,
"grad_norm": 0.28745728731155396,
"learning_rate": 0.00019510565162951537,
"loss": 0.9031,
"step": 47
},
{
"epoch": 0.023473317439941317,
"grad_norm": 0.558315098285675,
"learning_rate": 0.00019451838281608197,
"loss": 1.0545,
"step": 48
},
{
"epoch": 0.02396234488660676,
"grad_norm": 0.5743781924247742,
"learning_rate": 0.00019389883606150566,
"loss": 1.0951,
"step": 49
},
{
"epoch": 0.024451372333272204,
"grad_norm": 0.41075870394706726,
"learning_rate": 0.00019324722294043558,
"loss": 1.0124,
"step": 50
},
{
"epoch": 0.024940399779937648,
"grad_norm": 0.2487930953502655,
"learning_rate": 0.00019256376597815564,
"loss": 0.9894,
"step": 51
},
{
"epoch": 0.02542942722660309,
"grad_norm": 0.39657795429229736,
"learning_rate": 0.00019184869857459232,
"loss": 1.138,
"step": 52
},
{
"epoch": 0.02591845467326854,
"grad_norm": 0.5362446308135986,
"learning_rate": 0.00019110226492460885,
"loss": 1.0342,
"step": 53
},
{
"epoch": 0.026407482119933982,
"grad_norm": 0.5089200139045715,
"learning_rate": 0.0001903247199346129,
"loss": 1.1193,
"step": 54
},
{
"epoch": 0.026896509566599426,
"grad_norm": 0.28259485960006714,
"learning_rate": 0.00018951632913550626,
"loss": 0.9054,
"step": 55
},
{
"epoch": 0.02738553701326487,
"grad_norm": 0.3576945662498474,
"learning_rate": 0.0001886773685920062,
"loss": 1.0002,
"step": 56
},
{
"epoch": 0.027874564459930314,
"grad_norm": 0.2915222942829132,
"learning_rate": 0.0001878081248083698,
"loss": 0.9491,
"step": 57
},
{
"epoch": 0.028363591906595757,
"grad_norm": 0.296265572309494,
"learning_rate": 0.00018690889463055283,
"loss": 1.0483,
"step": 58
},
{
"epoch": 0.0288526193532612,
"grad_norm": 0.3489420413970947,
"learning_rate": 0.00018597998514483725,
"loss": 0.9064,
"step": 59
},
{
"epoch": 0.029341646799926645,
"grad_norm": 0.37772810459136963,
"learning_rate": 0.00018502171357296144,
"loss": 0.9779,
"step": 60
},
{
"epoch": 0.02983067424659209,
"grad_norm": 0.5371522903442383,
"learning_rate": 0.00018403440716378928,
"loss": 1.0268,
"step": 61
},
{
"epoch": 0.030319701693257536,
"grad_norm": 0.3227771520614624,
"learning_rate": 0.00018301840308155507,
"loss": 1.0642,
"step": 62
},
{
"epoch": 0.03080872913992298,
"grad_norm": 0.4422720670700073,
"learning_rate": 0.00018197404829072215,
"loss": 1.1083,
"step": 63
},
{
"epoch": 0.03129775658658842,
"grad_norm": 0.5397651791572571,
"learning_rate": 0.00018090169943749476,
"loss": 1.2271,
"step": 64
},
{
"epoch": 0.03178678403325386,
"grad_norm": 0.4943908452987671,
"learning_rate": 0.000179801722728024,
"loss": 0.9873,
"step": 65
},
{
"epoch": 0.03227581147991931,
"grad_norm": 0.31046292185783386,
"learning_rate": 0.00017867449380334834,
"loss": 0.9235,
"step": 66
},
{
"epoch": 0.03276483892658476,
"grad_norm": 0.4786432981491089,
"learning_rate": 0.00017752039761111297,
"loss": 1.0804,
"step": 67
},
{
"epoch": 0.0332538663732502,
"grad_norm": 0.38078072667121887,
"learning_rate": 0.00017633982827411032,
"loss": 1.0508,
"step": 68
},
{
"epoch": 0.0332538663732502,
"eval_loss": 1.0058449506759644,
"eval_runtime": 1360.1032,
"eval_samples_per_second": 1.899,
"eval_steps_per_second": 0.633,
"step": 68
},
{
"epoch": 0.033742893819915645,
"grad_norm": 0.302775114774704,
"learning_rate": 0.00017513318895568737,
"loss": 1.0186,
"step": 69
},
{
"epoch": 0.03423192126658109,
"grad_norm": 0.7890788316726685,
"learning_rate": 0.00017390089172206592,
"loss": 0.9441,
"step": 70
},
{
"epoch": 0.03472094871324653,
"grad_norm": 0.3507513999938965,
"learning_rate": 0.00017264335740162242,
"loss": 0.9604,
"step": 71
},
{
"epoch": 0.035209976159911976,
"grad_norm": 0.39182308316230774,
"learning_rate": 0.00017136101544117525,
"loss": 1.0598,
"step": 72
},
{
"epoch": 0.03569900360657742,
"grad_norm": 0.3877043128013611,
"learning_rate": 0.0001700543037593291,
"loss": 0.8719,
"step": 73
},
{
"epoch": 0.036188031053242864,
"grad_norm": 0.6628935933113098,
"learning_rate": 0.00016872366859692627,
"loss": 0.9241,
"step": 74
},
{
"epoch": 0.03667705849990831,
"grad_norm": 0.45674100518226624,
"learning_rate": 0.00016736956436465573,
"loss": 1.2833,
"step": 75
},
{
"epoch": 0.03716608594657375,
"grad_norm": 0.5053292512893677,
"learning_rate": 0.0001659924534878723,
"loss": 0.9678,
"step": 76
},
{
"epoch": 0.037655113393239195,
"grad_norm": 0.31924471259117126,
"learning_rate": 0.00016459280624867874,
"loss": 0.9702,
"step": 77
},
{
"epoch": 0.03814414083990464,
"grad_norm": 0.37664365768432617,
"learning_rate": 0.0001631711006253251,
"loss": 0.8382,
"step": 78
},
{
"epoch": 0.03863316828657008,
"grad_norm": 0.3583241403102875,
"learning_rate": 0.0001617278221289793,
"loss": 1.0149,
"step": 79
},
{
"epoch": 0.039122195733235526,
"grad_norm": 0.3891322910785675,
"learning_rate": 0.00016026346363792567,
"loss": 1.2247,
"step": 80
},
{
"epoch": 0.03961122317990097,
"grad_norm": 0.32489892840385437,
"learning_rate": 0.00015877852522924732,
"loss": 0.9108,
"step": 81
},
{
"epoch": 0.040100250626566414,
"grad_norm": 0.3395017981529236,
"learning_rate": 0.00015727351400805052,
"loss": 0.9968,
"step": 82
},
{
"epoch": 0.04058927807323186,
"grad_norm": 0.34815898537635803,
"learning_rate": 0.00015574894393428855,
"loss": 0.967,
"step": 83
},
{
"epoch": 0.0410783055198973,
"grad_norm": 0.4931004047393799,
"learning_rate": 0.00015420533564724495,
"loss": 1.0232,
"step": 84
},
{
"epoch": 0.041567332966562745,
"grad_norm": 0.45658788084983826,
"learning_rate": 0.0001526432162877356,
"loss": 1.0845,
"step": 85
},
{
"epoch": 0.042056360413228196,
"grad_norm": 0.2645609378814697,
"learning_rate": 0.0001510631193180907,
"loss": 0.8387,
"step": 86
},
{
"epoch": 0.04254538785989364,
"grad_norm": 0.2979009747505188,
"learning_rate": 0.0001494655843399779,
"loss": 1.0789,
"step": 87
},
{
"epoch": 0.04303441530655908,
"grad_norm": 0.3363690674304962,
"learning_rate": 0.00014785115691012864,
"loss": 0.9624,
"step": 88
},
{
"epoch": 0.04352344275322453,
"grad_norm": 0.700744092464447,
"learning_rate": 0.00014622038835403133,
"loss": 1.0248,
"step": 89
},
{
"epoch": 0.04401247019988997,
"grad_norm": 0.4093436300754547,
"learning_rate": 0.00014457383557765386,
"loss": 0.993,
"step": 90
}
],
"logging_steps": 1,
"max_steps": 200,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 5,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.995058775995187e+17,
"train_batch_size": 3,
"trial_name": null,
"trial_params": null
}