Models_Ouput / last-checkpoint /trainer_state.json
youssefedweqd's picture
Training in progress, step 4900, checkpoint
623a887 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.8112582781456954,
"eval_steps": 100,
"global_step": 4900,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.008278145695364239,
"grad_norm": 2.812814950942993,
"learning_rate": 2.7041942604856512e-06,
"loss": 1.4882,
"step": 50
},
{
"epoch": 0.016556291390728478,
"grad_norm": 2.6125919818878174,
"learning_rate": 5.463576158940398e-06,
"loss": 1.2862,
"step": 100
},
{
"epoch": 0.016556291390728478,
"eval_loss": 1.2020893096923828,
"eval_runtime": 1896.4557,
"eval_samples_per_second": 3.185,
"eval_steps_per_second": 3.185,
"step": 100
},
{
"epoch": 0.024834437086092714,
"grad_norm": 2.393519878387451,
"learning_rate": 8.222958057395145e-06,
"loss": 1.1353,
"step": 150
},
{
"epoch": 0.033112582781456956,
"grad_norm": 2.5604305267333984,
"learning_rate": 1.0982339955849891e-05,
"loss": 1.0539,
"step": 200
},
{
"epoch": 0.033112582781456956,
"eval_loss": 1.0201358795166016,
"eval_runtime": 1896.3026,
"eval_samples_per_second": 3.185,
"eval_steps_per_second": 3.185,
"step": 200
},
{
"epoch": 0.041390728476821195,
"grad_norm": 3.0053958892822266,
"learning_rate": 1.3741721854304637e-05,
"loss": 0.9816,
"step": 250
},
{
"epoch": 0.04966887417218543,
"grad_norm": 3.028010368347168,
"learning_rate": 1.6501103752759385e-05,
"loss": 0.9372,
"step": 300
},
{
"epoch": 0.057947019867549666,
"grad_norm": 2.484466552734375,
"learning_rate": 1.926048565121413e-05,
"loss": 0.9137,
"step": 350
},
{
"epoch": 0.06622516556291391,
"grad_norm": 3.078425884246826,
"learning_rate": 2.2019867549668874e-05,
"loss": 0.8786,
"step": 400
},
{
"epoch": 0.07450331125827815,
"grad_norm": 2.6672778129577637,
"learning_rate": 2.477924944812362e-05,
"loss": 0.8348,
"step": 450
},
{
"epoch": 0.08278145695364239,
"grad_norm": 2.8199753761291504,
"learning_rate": 2.753863134657837e-05,
"loss": 0.8417,
"step": 500
},
{
"epoch": 0.09105960264900662,
"grad_norm": 2.961965322494507,
"learning_rate": 3.0298013245033112e-05,
"loss": 0.8209,
"step": 550
},
{
"epoch": 0.09933774834437085,
"grad_norm": 2.5217514038085938,
"learning_rate": 3.305739514348786e-05,
"loss": 0.8235,
"step": 600
},
{
"epoch": 0.1076158940397351,
"grad_norm": 2.4740138053894043,
"learning_rate": 3.581677704194261e-05,
"loss": 0.7918,
"step": 650
},
{
"epoch": 0.11589403973509933,
"grad_norm": 2.5529448986053467,
"learning_rate": 3.8576158940397354e-05,
"loss": 0.7749,
"step": 700
},
{
"epoch": 0.12417218543046357,
"grad_norm": 2.151698589324951,
"learning_rate": 4.13355408388521e-05,
"loss": 0.7838,
"step": 750
},
{
"epoch": 0.13245033112582782,
"grad_norm": 1.9026315212249756,
"learning_rate": 4.4094922737306846e-05,
"loss": 0.7668,
"step": 800
},
{
"epoch": 0.14072847682119205,
"grad_norm": 1.9555529356002808,
"learning_rate": 4.685430463576159e-05,
"loss": 0.7684,
"step": 850
},
{
"epoch": 0.1490066225165563,
"grad_norm": 2.308894157409668,
"learning_rate": 4.961368653421634e-05,
"loss": 0.7661,
"step": 900
},
{
"epoch": 0.15728476821192053,
"grad_norm": 2.3362715244293213,
"learning_rate": 5.237306843267108e-05,
"loss": 0.736,
"step": 950
},
{
"epoch": 0.16556291390728478,
"grad_norm": 1.8228410482406616,
"learning_rate": 5.513245033112583e-05,
"loss": 0.7213,
"step": 1000
},
{
"epoch": 0.173841059602649,
"grad_norm": 1.8289754390716553,
"learning_rate": 5.789183222958058e-05,
"loss": 0.7335,
"step": 1050
},
{
"epoch": 0.18211920529801323,
"grad_norm": 1.4989681243896484,
"learning_rate": 6.065121412803533e-05,
"loss": 0.7326,
"step": 1100
},
{
"epoch": 0.19039735099337748,
"grad_norm": 1.5326098203659058,
"learning_rate": 6.341059602649006e-05,
"loss": 0.7311,
"step": 1150
},
{
"epoch": 0.1986754966887417,
"grad_norm": 1.4897147417068481,
"learning_rate": 6.616997792494481e-05,
"loss": 0.6918,
"step": 1200
},
{
"epoch": 0.20695364238410596,
"grad_norm": 1.634765863418579,
"learning_rate": 6.892935982339957e-05,
"loss": 0.7051,
"step": 1250
},
{
"epoch": 0.2152317880794702,
"grad_norm": 1.4463587999343872,
"learning_rate": 7.168874172185431e-05,
"loss": 0.6955,
"step": 1300
},
{
"epoch": 0.22350993377483444,
"grad_norm": 1.632133960723877,
"learning_rate": 7.444812362030905e-05,
"loss": 0.6901,
"step": 1350
},
{
"epoch": 0.23178807947019867,
"grad_norm": 1.4062328338623047,
"learning_rate": 7.72075055187638e-05,
"loss": 0.6833,
"step": 1400
},
{
"epoch": 0.24006622516556292,
"grad_norm": 1.2914466857910156,
"learning_rate": 7.996688741721855e-05,
"loss": 0.6663,
"step": 1450
},
{
"epoch": 0.24834437086092714,
"grad_norm": 1.4995919466018677,
"learning_rate": 8.272626931567329e-05,
"loss": 0.6959,
"step": 1500
},
{
"epoch": 0.25662251655629137,
"grad_norm": 1.1299749612808228,
"learning_rate": 8.548565121412803e-05,
"loss": 0.6685,
"step": 1550
},
{
"epoch": 0.26490066225165565,
"grad_norm": 1.329004168510437,
"learning_rate": 8.824503311258279e-05,
"loss": 0.6678,
"step": 1600
},
{
"epoch": 0.2731788079470199,
"grad_norm": 1.5191948413848877,
"learning_rate": 9.100441501103754e-05,
"loss": 0.6731,
"step": 1650
},
{
"epoch": 0.2814569536423841,
"grad_norm": 1.739169716835022,
"learning_rate": 9.376379690949227e-05,
"loss": 0.6691,
"step": 1700
},
{
"epoch": 0.2897350993377483,
"grad_norm": 1.2906118631362915,
"learning_rate": 9.652317880794703e-05,
"loss": 0.6718,
"step": 1750
},
{
"epoch": 0.2980132450331126,
"grad_norm": 1.289502501487732,
"learning_rate": 9.928256070640178e-05,
"loss": 0.6581,
"step": 1800
},
{
"epoch": 0.30629139072847683,
"grad_norm": 1.3923128843307495,
"learning_rate": 9.999872989402833e-05,
"loss": 0.6589,
"step": 1850
},
{
"epoch": 0.31456953642384106,
"grad_norm": 1.1048816442489624,
"learning_rate": 9.999297790520483e-05,
"loss": 0.6341,
"step": 1900
},
{
"epoch": 0.3228476821192053,
"grad_norm": 1.3568603992462158,
"learning_rate": 9.998258777484084e-05,
"loss": 0.6318,
"step": 1950
},
{
"epoch": 0.33112582781456956,
"grad_norm": 0.923786997795105,
"learning_rate": 9.996756046688961e-05,
"loss": 0.6318,
"step": 2000
},
{
"epoch": 0.3394039735099338,
"grad_norm": 1.102367877960205,
"learning_rate": 9.994789737552259e-05,
"loss": 0.6193,
"step": 2050
},
{
"epoch": 0.347682119205298,
"grad_norm": 1.0738896131515503,
"learning_rate": 9.992360032500001e-05,
"loss": 0.6184,
"step": 2100
},
{
"epoch": 0.35596026490066224,
"grad_norm": 1.279288649559021,
"learning_rate": 9.98946715695016e-05,
"loss": 0.626,
"step": 2150
},
{
"epoch": 0.36423841059602646,
"grad_norm": 1.2009036540985107,
"learning_rate": 9.986111379291759e-05,
"loss": 0.6305,
"step": 2200
},
{
"epoch": 0.37251655629139074,
"grad_norm": 0.8177038431167603,
"learning_rate": 9.982293010859955e-05,
"loss": 0.6266,
"step": 2250
},
{
"epoch": 0.38079470198675497,
"grad_norm": 1.2464983463287354,
"learning_rate": 9.978012405907165e-05,
"loss": 0.6148,
"step": 2300
},
{
"epoch": 0.3890728476821192,
"grad_norm": 1.2841860055923462,
"learning_rate": 9.973269961570195e-05,
"loss": 0.5946,
"step": 2350
},
{
"epoch": 0.3973509933774834,
"grad_norm": 1.2200431823730469,
"learning_rate": 9.968066117833401e-05,
"loss": 0.6166,
"step": 2400
},
{
"epoch": 0.4056291390728477,
"grad_norm": 1.128247857093811,
"learning_rate": 9.962401357487863e-05,
"loss": 0.5992,
"step": 2450
},
{
"epoch": 0.4139072847682119,
"grad_norm": 1.0683091878890991,
"learning_rate": 9.956276206086597e-05,
"loss": 0.6048,
"step": 2500
},
{
"epoch": 0.42218543046357615,
"grad_norm": 1.1819758415222168,
"learning_rate": 9.949691231895791e-05,
"loss": 0.5944,
"step": 2550
},
{
"epoch": 0.4304635761589404,
"grad_norm": 1.0043411254882812,
"learning_rate": 9.942647045842095e-05,
"loss": 0.5962,
"step": 2600
},
{
"epoch": 0.43874172185430466,
"grad_norm": 1.0588668584823608,
"learning_rate": 9.93514430145593e-05,
"loss": 0.6067,
"step": 2650
},
{
"epoch": 0.4470198675496689,
"grad_norm": 0.9364084601402283,
"learning_rate": 9.927183694810862e-05,
"loss": 0.5928,
"step": 2700
},
{
"epoch": 0.4552980132450331,
"grad_norm": 1.155172348022461,
"learning_rate": 9.918765964459022e-05,
"loss": 0.5987,
"step": 2750
},
{
"epoch": 0.46357615894039733,
"grad_norm": 1.1639224290847778,
"learning_rate": 9.909891891362587e-05,
"loss": 0.5745,
"step": 2800
},
{
"epoch": 0.4718543046357616,
"grad_norm": 0.9658174514770508,
"learning_rate": 9.900562298821323e-05,
"loss": 0.5825,
"step": 2850
},
{
"epoch": 0.48013245033112584,
"grad_norm": 1.118033766746521,
"learning_rate": 9.890778052396205e-05,
"loss": 0.5806,
"step": 2900
},
{
"epoch": 0.48841059602649006,
"grad_norm": 0.9781912565231323,
"learning_rate": 9.880540059829115e-05,
"loss": 0.5712,
"step": 2950
},
{
"epoch": 0.4966887417218543,
"grad_norm": 1.2145684957504272,
"learning_rate": 9.869849270958622e-05,
"loss": 0.5855,
"step": 3000
},
{
"epoch": 0.5049668874172185,
"grad_norm": 0.999279260635376,
"learning_rate": 9.858706677631862e-05,
"loss": 0.5843,
"step": 3050
},
{
"epoch": 0.5132450331125827,
"grad_norm": 1.098258137702942,
"learning_rate": 9.847113313612517e-05,
"loss": 0.5605,
"step": 3100
},
{
"epoch": 0.5215231788079471,
"grad_norm": 0.627949059009552,
"learning_rate": 9.835070254484912e-05,
"loss": 0.5538,
"step": 3150
},
{
"epoch": 0.5298013245033113,
"grad_norm": 1.0991902351379395,
"learning_rate": 9.822578617554219e-05,
"loss": 0.5555,
"step": 3200
},
{
"epoch": 0.5380794701986755,
"grad_norm": 0.9670843482017517,
"learning_rate": 9.8096395617428e-05,
"loss": 0.5647,
"step": 3250
},
{
"epoch": 0.5463576158940397,
"grad_norm": 0.9838133454322815,
"learning_rate": 9.796254287482693e-05,
"loss": 0.5561,
"step": 3300
},
{
"epoch": 0.554635761589404,
"grad_norm": 1.1465744972229004,
"learning_rate": 9.782424036604234e-05,
"loss": 0.559,
"step": 3350
},
{
"epoch": 0.5629139072847682,
"grad_norm": 1.1423758268356323,
"learning_rate": 9.768150092220849e-05,
"loss": 0.5517,
"step": 3400
},
{
"epoch": 0.5711920529801324,
"grad_norm": 1.1365066766738892,
"learning_rate": 9.753433778610008e-05,
"loss": 0.5464,
"step": 3450
},
{
"epoch": 0.5794701986754967,
"grad_norm": 0.81045001745224,
"learning_rate": 9.738276461090371e-05,
"loss": 0.5493,
"step": 3500
},
{
"epoch": 0.5877483443708609,
"grad_norm": 1.0236687660217285,
"learning_rate": 9.72267954589511e-05,
"loss": 0.567,
"step": 3550
},
{
"epoch": 0.5960264900662252,
"grad_norm": 0.9495602250099182,
"learning_rate": 9.706644480041455e-05,
"loss": 0.5474,
"step": 3600
},
{
"epoch": 0.6043046357615894,
"grad_norm": 0.960738480091095,
"learning_rate": 9.690172751196437e-05,
"loss": 0.5238,
"step": 3650
},
{
"epoch": 0.6125827814569537,
"grad_norm": 1.0488675832748413,
"learning_rate": 9.67326588753887e-05,
"loss": 0.521,
"step": 3700
},
{
"epoch": 0.6208609271523179,
"grad_norm": 0.8753538727760315,
"learning_rate": 9.65592545761758e-05,
"loss": 0.5232,
"step": 3750
},
{
"epoch": 0.6291390728476821,
"grad_norm": 1.0551217794418335,
"learning_rate": 9.638153070205871e-05,
"loss": 0.5432,
"step": 3800
},
{
"epoch": 0.6374172185430463,
"grad_norm": 1.158676028251648,
"learning_rate": 9.619950374152278e-05,
"loss": 0.5416,
"step": 3850
},
{
"epoch": 0.6456953642384106,
"grad_norm": 1.0036752223968506,
"learning_rate": 9.601319058227589e-05,
"loss": 0.5496,
"step": 3900
},
{
"epoch": 0.6539735099337748,
"grad_norm": 0.8905594348907471,
"learning_rate": 9.58226085096817e-05,
"loss": 0.5335,
"step": 3950
},
{
"epoch": 0.6622516556291391,
"grad_norm": 0.9868190884590149,
"learning_rate": 9.562777520515598e-05,
"loss": 0.5094,
"step": 4000
},
{
"epoch": 0.6705298013245033,
"grad_norm": 0.9672690629959106,
"learning_rate": 9.542870874452618e-05,
"loss": 0.5061,
"step": 4050
},
{
"epoch": 0.6788079470198676,
"grad_norm": 1.044123888015747,
"learning_rate": 9.52254275963545e-05,
"loss": 0.5253,
"step": 4100
},
{
"epoch": 0.6870860927152318,
"grad_norm": 1.0346958637237549,
"learning_rate": 9.501795062022434e-05,
"loss": 0.5149,
"step": 4150
},
{
"epoch": 0.695364238410596,
"grad_norm": 1.0799248218536377,
"learning_rate": 9.48062970649907e-05,
"loss": 0.5207,
"step": 4200
},
{
"epoch": 0.7036423841059603,
"grad_norm": 0.9847925901412964,
"learning_rate": 9.459048656699427e-05,
"loss": 0.531,
"step": 4250
},
{
"epoch": 0.7119205298013245,
"grad_norm": 1.134179949760437,
"learning_rate": 9.43705391482397e-05,
"loss": 0.5202,
"step": 4300
},
{
"epoch": 0.7201986754966887,
"grad_norm": 0.9750307202339172,
"learning_rate": 9.414647521453798e-05,
"loss": 0.5183,
"step": 4350
},
{
"epoch": 0.7284768211920529,
"grad_norm": 1.372010350227356,
"learning_rate": 9.391831555361341e-05,
"loss": 0.5203,
"step": 4400
},
{
"epoch": 0.7367549668874173,
"grad_norm": 0.9671643376350403,
"learning_rate": 9.36860813331748e-05,
"loss": 0.5313,
"step": 4450
},
{
"epoch": 0.7450331125827815,
"grad_norm": 1.270264983177185,
"learning_rate": 9.344979409895178e-05,
"loss": 0.5236,
"step": 4500
},
{
"epoch": 0.7533112582781457,
"grad_norm": 1.1816293001174927,
"learning_rate": 9.320947577269581e-05,
"loss": 0.518,
"step": 4550
},
{
"epoch": 0.7615894039735099,
"grad_norm": 0.8809527158737183,
"learning_rate": 9.29651486501464e-05,
"loss": 0.5086,
"step": 4600
},
{
"epoch": 0.7698675496688742,
"grad_norm": 0.9570929408073425,
"learning_rate": 9.271683539896257e-05,
"loss": 0.5195,
"step": 4650
},
{
"epoch": 0.7781456953642384,
"grad_norm": 1.147157907485962,
"learning_rate": 9.246455905661983e-05,
"loss": 0.509,
"step": 4700
},
{
"epoch": 0.7864238410596026,
"grad_norm": 0.9548070430755615,
"learning_rate": 9.220834302827295e-05,
"loss": 0.5078,
"step": 4750
},
{
"epoch": 0.7947019867549668,
"grad_norm": 1.0823866128921509,
"learning_rate": 9.194821108458438e-05,
"loss": 0.5088,
"step": 4800
},
{
"epoch": 0.8029801324503312,
"grad_norm": 1.0077965259552002,
"learning_rate": 9.168418735951902e-05,
"loss": 0.4994,
"step": 4850
},
{
"epoch": 0.8112582781456954,
"grad_norm": 1.1390560865402222,
"learning_rate": 9.141629634810516e-05,
"loss": 0.5098,
"step": 4900
}
],
"logging_steps": 50,
"max_steps": 18120,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.842256577599898e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}