TamilRAV / trainer_state.json
Meclin's picture
Upload folder using huggingface_hub
ef84600 verified
Raw
History Blame Contribute Delete
13 kB
{
"best_global_step": 650,
"best_metric": 0.6772647500038147,
"best_model_checkpoint": "./tamilrav_checkpoints/checkpoint-650",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 650,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.015396458814472672,
"grad_norm": 0.29363131523132324,
"learning_rate": 1.8e-05,
"loss": 1.4163263320922852,
"step": 10
},
{
"epoch": 0.030792917628945343,
"grad_norm": 0.265934020280838,
"learning_rate": 3.8e-05,
"loss": 1.188729953765869,
"step": 20
},
{
"epoch": 0.046189376443418015,
"grad_norm": 0.1696174591779709,
"learning_rate": 5.8e-05,
"loss": 1.0048276901245117,
"step": 30
},
{
"epoch": 0.061585835257890686,
"grad_norm": 0.1621861755847931,
"learning_rate": 7.800000000000001e-05,
"loss": 0.939247989654541,
"step": 40
},
{
"epoch": 0.07698229407236336,
"grad_norm": 0.16908901929855347,
"learning_rate": 9.8e-05,
"loss": 0.9107258796691895,
"step": 50
},
{
"epoch": 0.09237875288683603,
"grad_norm": 0.16060316562652588,
"learning_rate": 0.000118,
"loss": 0.8661402702331543,
"step": 60
},
{
"epoch": 0.1077752117013087,
"grad_norm": 0.15463921427726746,
"learning_rate": 0.000138,
"loss": 0.8585456848144531,
"step": 70
},
{
"epoch": 0.12317167051578137,
"grad_norm": 0.14651408791542053,
"learning_rate": 0.00015800000000000002,
"loss": 0.8524063110351563,
"step": 80
},
{
"epoch": 0.13856812933025403,
"grad_norm": 0.1593480259180069,
"learning_rate": 0.00017800000000000002,
"loss": 0.8573471069335937,
"step": 90
},
{
"epoch": 0.15396458814472672,
"grad_norm": 0.14837513864040375,
"learning_rate": 0.00019800000000000002,
"loss": 0.8089125633239747,
"step": 100
},
{
"epoch": 0.16936104695919937,
"grad_norm": 0.1440625786781311,
"learning_rate": 0.00019998832107530538,
"loss": 0.8235550880432129,
"step": 110
},
{
"epoch": 0.18475750577367206,
"grad_norm": 0.1411304473876953,
"learning_rate": 0.00019994795298602624,
"loss": 0.8093386650085449,
"step": 120
},
{
"epoch": 0.20015396458814472,
"grad_norm": 0.1452978402376175,
"learning_rate": 0.0001998787631859678,
"loss": 0.8320710182189941,
"step": 130
},
{
"epoch": 0.2155504234026174,
"grad_norm": 0.13940446078777313,
"learning_rate": 0.00019978077162719857,
"loss": 0.821429443359375,
"step": 140
},
{
"epoch": 0.23094688221709006,
"grad_norm": 0.14506584405899048,
"learning_rate": 0.00019965400656726942,
"loss": 0.7923899173736573,
"step": 150
},
{
"epoch": 0.24634334103156275,
"grad_norm": 0.14369885623455048,
"learning_rate": 0.000199498504561065,
"loss": 0.7901590824127197,
"step": 160
},
{
"epoch": 0.2617397998460354,
"grad_norm": 0.14093971252441406,
"learning_rate": 0.00019931431045026255,
"loss": 0.8111134529113769,
"step": 170
},
{
"epoch": 0.27713625866050806,
"grad_norm": 0.13777726888656616,
"learning_rate": 0.00019910147735040102,
"loss": 0.791386604309082,
"step": 180
},
{
"epoch": 0.2925327174749808,
"grad_norm": 0.1432289332151413,
"learning_rate": 0.00019886006663556426,
"loss": 0.7923940181732178,
"step": 190
},
{
"epoch": 0.30792917628945343,
"grad_norm": 0.13564977049827576,
"learning_rate": 0.00019859014792068282,
"loss": 0.7843753337860108,
"step": 200
},
{
"epoch": 0.3233256351039261,
"grad_norm": 0.13792838156223297,
"learning_rate": 0.00019829179904145918,
"loss": 0.7822946071624756,
"step": 210
},
{
"epoch": 0.33872209391839875,
"grad_norm": 0.13952118158340454,
"learning_rate": 0.00019796510603192253,
"loss": 0.7925955772399902,
"step": 220
},
{
"epoch": 0.35411855273287146,
"grad_norm": 0.13627943396568298,
"learning_rate": 0.00019761016309961948,
"loss": 0.7559749603271484,
"step": 230
},
{
"epoch": 0.3695150115473441,
"grad_norm": 0.13699838519096375,
"learning_rate": 0.00019722707259844756,
"loss": 0.769603681564331,
"step": 240
},
{
"epoch": 0.3849114703618168,
"grad_norm": 0.13574793934822083,
"learning_rate": 0.0001968159449991397,
"loss": 0.7658140659332275,
"step": 250
},
{
"epoch": 0.40030792917628943,
"grad_norm": 0.13491615653038025,
"learning_rate": 0.00019637689885740827,
"loss": 0.7487314701080322,
"step": 260
},
{
"epoch": 0.41570438799076215,
"grad_norm": 0.1336905062198639,
"learning_rate": 0.0001959100607797573,
"loss": 0.7576127052307129,
"step": 270
},
{
"epoch": 0.4311008468052348,
"grad_norm": 0.12657032907009125,
"learning_rate": 0.0001954155653869735,
"loss": 0.7317625999450683,
"step": 280
},
{
"epoch": 0.44649730561970746,
"grad_norm": 0.13930626213550568,
"learning_rate": 0.00019489355527530576,
"loss": 0.7514139652252197,
"step": 290
},
{
"epoch": 0.4618937644341801,
"grad_norm": 0.1464177966117859,
"learning_rate": 0.00019434418097534554,
"loss": 0.750247573852539,
"step": 300
},
{
"epoch": 0.47729022324865283,
"grad_norm": 0.14013901352882385,
"learning_rate": 0.00019376760090861824,
"loss": 0.736499834060669,
"step": 310
},
{
"epoch": 0.4926866820631255,
"grad_norm": 0.131071999669075,
"learning_rate": 0.0001931639813419,
"loss": 0.7452194690704346,
"step": 320
},
{
"epoch": 0.5080831408775982,
"grad_norm": 0.13927963376045227,
"learning_rate": 0.0001925334963392717,
"loss": 0.7342241764068603,
"step": 330
},
{
"epoch": 0.5234795996920708,
"grad_norm": 0.1341823786497116,
"learning_rate": 0.00019187632771192466,
"loss": 0.7511595726013184,
"step": 340
},
{
"epoch": 0.5388760585065435,
"grad_norm": 0.13563531637191772,
"learning_rate": 0.00019119266496573208,
"loss": 0.7299445629119873,
"step": 350
},
{
"epoch": 0.5542725173210161,
"grad_norm": 0.1313074827194214,
"learning_rate": 0.00019048270524660196,
"loss": 0.7165977001190186,
"step": 360
},
{
"epoch": 0.5696689761354888,
"grad_norm": 0.14486627280712128,
"learning_rate": 0.00018974665328362654,
"loss": 0.710158109664917,
"step": 370
},
{
"epoch": 0.5850654349499615,
"grad_norm": 0.14865578711032867,
"learning_rate": 0.00018898472133004531,
"loss": 0.7281487941741943,
"step": 380
},
{
"epoch": 0.6004618937644342,
"grad_norm": 0.14025066792964935,
"learning_rate": 0.00018819712910203795,
"loss": 0.7306881904602051,
"step": 390
},
{
"epoch": 0.6158583525789069,
"grad_norm": 0.1385616809129715,
"learning_rate": 0.00018738410371536545,
"loss": 0.715015983581543,
"step": 400
},
{
"epoch": 0.6312548113933796,
"grad_norm": 0.13913969695568085,
"learning_rate": 0.00018654587961987754,
"loss": 0.6998976230621338,
"step": 410
},
{
"epoch": 0.6466512702078522,
"grad_norm": 0.14493151009082794,
"learning_rate": 0.00018568269853190482,
"loss": 0.7252633094787597,
"step": 420
},
{
"epoch": 0.6620477290223249,
"grad_norm": 0.1407926082611084,
"learning_rate": 0.00018479480936455577,
"loss": 0.705822229385376,
"step": 430
},
{
"epoch": 0.6774441878367975,
"grad_norm": 0.1485988348722458,
"learning_rate": 0.00018388246815593848,
"loss": 0.7083362579345703,
"step": 440
},
{
"epoch": 0.6928406466512702,
"grad_norm": 0.13596095144748688,
"learning_rate": 0.00018294593799532752,
"loss": 0.7086032867431641,
"step": 450
},
{
"epoch": 0.7082371054657429,
"grad_norm": 0.1445472687482834,
"learning_rate": 0.00018198548894729765,
"loss": 0.7101798057556152,
"step": 460
},
{
"epoch": 0.7236335642802155,
"grad_norm": 0.13956452906131744,
"learning_rate": 0.0001810013979738462,
"loss": 0.6983530044555664,
"step": 470
},
{
"epoch": 0.7390300230946882,
"grad_norm": 0.14123016595840454,
"learning_rate": 0.00017999394885452605,
"loss": 0.6674805641174316,
"step": 480
},
{
"epoch": 0.7544264819091608,
"grad_norm": 0.1448190063238144,
"learning_rate": 0.00017896343210461319,
"loss": 0.68931565284729,
"step": 490
},
{
"epoch": 0.7698229407236336,
"grad_norm": 0.14515918493270874,
"learning_rate": 0.00017791014489133147,
"loss": 0.6919535636901856,
"step": 500
},
{
"epoch": 0.7852193995381063,
"grad_norm": 0.13914334774017334,
"learning_rate": 0.00017683439094815938,
"loss": 0.6638880729675293,
"step": 510
},
{
"epoch": 0.8006158583525789,
"grad_norm": 0.14346542954444885,
"learning_rate": 0.00017573648048724327,
"loss": 0.6776976585388184,
"step": 520
},
{
"epoch": 0.8160123171670516,
"grad_norm": 0.1400802731513977,
"learning_rate": 0.00017461673010994236,
"loss": 0.7085250854492188,
"step": 530
},
{
"epoch": 0.8314087759815243,
"grad_norm": 0.1518496870994568,
"learning_rate": 0.0001734754627155313,
"loss": 0.6884902000427247,
"step": 540
},
{
"epoch": 0.8468052347959969,
"grad_norm": 0.14645785093307495,
"learning_rate": 0.00017231300740808654,
"loss": 0.7027451992034912,
"step": 550
},
{
"epoch": 0.8622016936104696,
"grad_norm": 0.14037857949733734,
"learning_rate": 0.00017112969940158359,
"loss": 0.6541213035583496,
"step": 560
},
{
"epoch": 0.8775981524249422,
"grad_norm": 0.14151780307292938,
"learning_rate": 0.0001699258799232323,
"loss": 0.6418449401855468,
"step": 570
},
{
"epoch": 0.8929946112394149,
"grad_norm": 0.14554472267627716,
"learning_rate": 0.00016870189611507813,
"loss": 0.6883684635162354,
"step": 580
},
{
"epoch": 0.9083910700538876,
"grad_norm": 0.14308355748653412,
"learning_rate": 0.00016745810093389766,
"loss": 0.6611432552337646,
"step": 590
},
{
"epoch": 0.9237875288683602,
"grad_norm": 0.153996542096138,
"learning_rate": 0.0001661948530494176,
"loss": 0.6723202705383301,
"step": 600
},
{
"epoch": 0.939183987682833,
"grad_norm": 0.14034205675125122,
"learning_rate": 0.00016491251674088613,
"loss": 0.6433265686035157,
"step": 610
},
{
"epoch": 0.9545804464973057,
"grad_norm": 0.15295536816120148,
"learning_rate": 0.00016361146179202666,
"loss": 0.6759867191314697,
"step": 620
},
{
"epoch": 0.9699769053117783,
"grad_norm": 0.14949378371238708,
"learning_rate": 0.00016229206338440447,
"loss": 0.6557295799255372,
"step": 630
},
{
"epoch": 0.985373364126251,
"grad_norm": 0.15767240524291992,
"learning_rate": 0.00016095470198923648,
"loss": 0.6578701972961426,
"step": 640
},
{
"epoch": 1.0,
"grad_norm": 0.3094935119152069,
"learning_rate": 0.00015959976325767607,
"loss": 0.656089973449707,
"step": 650
},
{
"epoch": 1.0,
"eval_loss": 0.6772647500038147,
"eval_runtime": 193.0976,
"eval_samples_per_second": 2.978,
"eval_steps_per_second": 0.373,
"step": 650
}
],
"logging_steps": 10,
"max_steps": 1950,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 4.665508132426875e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}