{ "iterations": 31400, "parameter_count": 730164480, "tokens_per_step": 262144, "processed_tokens": 8231321600, "elapsed_seconds": 271423.6548615545, "best_validation_loss": 3.712334632873535, "history": [ { "iteration": 500, "train_loss": 6.657673358917236, "validation_loss": 6.6738409996032715, "learning_rate": 0.0002499772703356698, "elapsed_seconds": 4271.6415923163295, "projected_total_hours": 74.5164144437404, "projected_total_cost_usd": 103.57781607679915 }, { "iteration": 1000, "train_loss": 5.294595718383789, "validation_loss": 5.350976467132568, "learning_rate": 0.0002497196662435132, "elapsed_seconds": 8576.508910141885, "projected_total_hours": 74.80621660512645, "projected_total_cost_usd": 103.98064108112575 }, { "iteration": 1500, "train_loss": 4.833080291748047, "validation_loss": 4.898420810699463, "learning_rate": 0.00024917584502227825, "elapsed_seconds": 12891.454528726637, "projected_total_hours": 74.9614207781512, "projected_total_cost_usd": 104.19637488163015 }, { "iteration": 2000, "train_loss": 4.620269775390625, "validation_loss": 4.684819221496582, "learning_rate": 0.0002483471936922074, "elapsed_seconds": 17209.910989679396, "projected_total_hours": 75.05433403832403, "projected_total_cost_usd": 104.32552431327039 }, { "iteration": 2500, "train_loss": 4.481897354125977, "validation_loss": 4.551168441772461, "learning_rate": 0.0002472358257349446, "elapsed_seconds": 21535.88317859918, "projected_total_hours": 75.13630353422381, "projected_total_cost_usd": 104.43946191257109 }, { "iteration": 3000, "train_loss": 4.3804826736450195, "validation_loss": 4.452548503875732, "learning_rate": 0.0002458445757030842, "elapsed_seconds": 25854.85624050349, "projected_total_hours": 75.17060055109349, "projected_total_cost_usd": 104.48713476601993 }, { "iteration": 3500, "train_loss": 4.3059492111206055, "validation_loss": 4.380916118621826, "learning_rate": 0.0002441769919906223, "elapsed_seconds": 30173.152176752687, "projected_total_hours": 75.19341098016146, "projected_total_cost_usd": 104.51884126242442 }, { "iteration": 4000, "train_loss": 4.242997169494629, "validation_loss": 4.31741189956665, "learning_rate": 0.00024223732778275053, "elapsed_seconds": 34493.88701976091, "projected_total_hours": 75.21583697364531, "projected_total_cost_usd": 104.55001339336697 }, { "iteration": 4500, "train_loss": 4.189329147338867, "validation_loss": 4.266604423522949, "learning_rate": 0.0002400305302080742, "elapsed_seconds": 38816.21281320602, "projected_total_hours": 75.23636310707833, "projected_total_cost_usd": 104.57854471883887 }, { "iteration": 5000, "train_loss": 4.1476969718933105, "validation_loss": 4.227367877960205, "learning_rate": 0.00023756222772092247, "elapsed_seconds": 43158.014198705554, "projected_total_hours": 75.28675810218637, "projected_total_cost_usd": 104.64859376203904 }, { "iteration": 5500, "train_loss": 4.109938621520996, "validation_loss": 4.190089225769043, "learning_rate": 0.00023483871574593183, "elapsed_seconds": 47477.29582044482, "projected_total_hours": 75.29227721020038, "projected_total_cost_usd": 104.65626532217853 }, { "iteration": 6000, "train_loss": 4.076807022094727, "validation_loss": 4.157365798950195, "learning_rate": 0.0002318669406215168, "elapsed_seconds": 51794.34619618952, "projected_total_hours": 75.29363289631254, "projected_total_cost_usd": 104.65814972587442 }, { "iteration": 6500, "train_loss": 4.042527675628662, "validation_loss": 4.129131317138672, "learning_rate": 0.0002286544818831799, "elapsed_seconds": 56113.59111300856, "projected_total_hours": 75.29772482685765, "projected_total_cost_usd": 104.66383750933213 }, { "iteration": 7000, "train_loss": 4.053408622741699, "validation_loss": 4.135789394378662, "learning_rate": 0.00022520953293184766, "elapsed_seconds": 60422.58037369698, "projected_total_hours": 75.28845332278115, "projected_total_cost_usd": 104.65095011866579 }, { "iteration": 7500, "train_loss": 3.9998574256896973, "validation_loss": 4.0869526863098145, "learning_rate": 0.0002215408801365379, "elapsed_seconds": 64723.61913320422, "projected_total_hours": 75.2711718808375, "projected_total_cost_usd": 104.62692891436411 }, { "iteration": 8000, "train_loss": 3.9767398834228516, "validation_loss": 4.06421422958374, "learning_rate": 0.00021765788042465737, "elapsed_seconds": 69042.05848243833, "projected_total_hours": 75.27502209543623, "projected_total_cost_usd": 104.63228071265635 }, { "iteration": 8500, "train_loss": 3.9555294513702393, "validation_loss": 4.046372890472412, "learning_rate": 0.00021357043741708606, "elapsed_seconds": 73355.51514386386, "projected_total_hours": 75.27330638945507, "projected_total_cost_usd": 104.62989588134255 }, { "iteration": 9000, "train_loss": 3.934286117553711, "validation_loss": 4.028558254241943, "learning_rate": 0.00020928897616891504, "elapsed_seconds": 77672.20504949987, "projected_total_hours": 75.27491477019431, "projected_total_cost_usd": 104.63213153057009 }, { "iteration": 9500, "train_loss": 3.9189178943634033, "validation_loss": 4.014111518859863, "learning_rate": 0.0002048244165802629, "elapsed_seconds": 81986.55616268516, "projected_total_hours": 75.27420653533082, "projected_total_cost_usd": 104.63114708410983 }, { "iteration": 10000, "train_loss": 3.896970272064209, "validation_loss": 3.996415138244629, "learning_rate": 0.00020018814554498553, "elapsed_seconds": 86303.21427984536, "projected_total_hours": 75.27558134408734, "projected_total_cost_usd": 104.6330580682814 }, { "iteration": 10500, "train_loss": 3.882066011428833, "validation_loss": 3.9830996990203857, "learning_rate": 0.00019539198790831518, "elapsed_seconds": 90618.27073761076, "projected_total_hours": 75.2754947397084, "projected_total_cost_usd": 104.63293768819467 }, { "iteration": 11000, "train_loss": 3.8664541244506836, "validation_loss": 3.9677159786224365, "learning_rate": 0.000190448176307501, "elapsed_seconds": 94942.73910321295, "projected_total_hours": 75.28287898588098, "projected_total_cost_usd": 104.64320179037456 }, { "iteration": 11500, "train_loss": 3.851578950881958, "validation_loss": 3.9544689655303955, "learning_rate": 0.0001853693199723728, "elapsed_seconds": 99264.58329997957, "projected_total_hours": 75.2876308120618, "projected_total_cost_usd": 104.6498068287659 }, { "iteration": 12000, "train_loss": 3.8364357948303223, "validation_loss": 3.9412567615509033, "learning_rate": 0.00018016837256540257, "elapsed_seconds": 103622.94289319962, "projected_total_hours": 75.31852793626084, "projected_total_cost_usd": 104.69275383140256 }, { "iteration": 12500, "train_loss": 3.8234899044036865, "validation_loss": 3.929971694946289, "learning_rate": 0.00017485859914328803, "elapsed_seconds": 107943.69804026932, "projected_total_hours": 75.32071374365458, "projected_total_cost_usd": 104.69579210367986 }, { "iteration": 13000, "train_loss": 3.8084628582000732, "validation_loss": 3.917842388153076, "learning_rate": 0.00016945354232432232, "elapsed_seconds": 112258.32612586766, "projected_total_hours": 75.3186205203471, "projected_total_cost_usd": 104.69288252328245 }, { "iteration": 13500, "train_loss": 3.7961957454681396, "validation_loss": 3.906834125518799, "learning_rate": 0.00016396698774784145, "elapsed_seconds": 116572.56532915682, "projected_total_hours": 75.31643109743877, "projected_total_cost_usd": 104.68983922543988 }, { "iteration": 14000, "train_loss": 3.7818515300750732, "validation_loss": 3.8985838890075684, "learning_rate": 0.0001584129289138439, "elapsed_seconds": 120888.49704167992, "projected_total_hours": 75.31545252199899, "projected_total_cost_usd": 104.68847900557859 }, { "iteration": 14500, "train_loss": 3.7692983150482178, "validation_loss": 3.889788866043091, "learning_rate": 0.00015280553149246053, "elapsed_seconds": 125203.99943609536, "projected_total_hours": 75.3142831856972, "projected_total_cost_usd": 104.68685362811911 }, { "iteration": 15000, "train_loss": 3.7596471309661865, "validation_loss": 3.879643201828003, "learning_rate": 0.00014715909719430296, "elapsed_seconds": 129519.89085417241, "projected_total_hours": 75.31341801520396, "projected_total_cost_usd": 104.6856510411335 }, { "iteration": 15500, "train_loss": 3.748656988143921, "validation_loss": 3.8702828884124756, "learning_rate": 0.00014148802729383982, "elapsed_seconds": 133837.80417640507, "projected_total_hours": 75.31374643618493, "projected_total_cost_usd": 104.68610754629704 }, { "iteration": 16000, "train_loss": 3.735333204269409, "validation_loss": 3.8588755130767822, "learning_rate": 0.00013580678589883435, "elapsed_seconds": 138154.23407498747, "projected_total_hours": 75.31324565893414, "projected_total_cost_usd": 104.68541146591843 }, { "iteration": 16500, "train_loss": 3.7218117713928223, "validation_loss": 3.8520736694335938, "learning_rate": 0.00013012986305952577, "elapsed_seconds": 142472.6207278967, "projected_total_hours": 75.31380961036963, "projected_total_cost_usd": 104.68619535841378 }, { "iteration": 17000, "train_loss": 3.7100892066955566, "validation_loss": 3.8433520793914795, "learning_rate": 0.00012447173781164324, "elapsed_seconds": 146782.17834874988, "projected_total_hours": 75.3098104599795, "projected_total_cost_usd": 104.6806365393715 }, { "iteration": 17500, "train_loss": 3.7014331817626953, "validation_loss": 3.8347644805908203, "learning_rate": 0.00011884684124751335, "elapsed_seconds": 151125.83998079598, "projected_total_hours": 75.3230377047142, "projected_total_cost_usd": 104.69902240955273 }, { "iteration": 18000, "train_loss": 3.6897997856140137, "validation_loss": 3.826525926589966, "learning_rate": 0.00011326951970944703, "elapsed_seconds": 155471.40847074986, "projected_total_hours": 75.33645410465348, "projected_total_cost_usd": 104.71767120546833 }, { "iteration": 18500, "train_loss": 3.679842472076416, "validation_loss": 3.819714307785034, "learning_rate": 0.0001077539981992822, "elapsed_seconds": 159820.10016273707, "projected_total_hours": 75.3506177944436, "projected_total_cost_usd": 104.73735873427658 }, { "iteration": 19000, "train_loss": 3.6696887016296387, "validation_loss": 3.8133578300476074, "learning_rate": 0.00010231434409740646, "elapsed_seconds": 164168.3918394372, "projected_total_hours": 75.3638523941276, "projected_total_cost_usd": 104.75575482783736 }, { "iteration": 19500, "train_loss": 3.661100149154663, "validation_loss": 3.805586576461792, "learning_rate": 9.696443128379405e-05, "elapsed_seconds": 168500.78201993555, "projected_total_hours": 75.3692956613387, "projected_total_cost_usd": 104.76332096926077 }, { "iteration": 20000, "train_loss": 3.6516928672790527, "validation_loss": 3.79829740524292, "learning_rate": 9.171790475256764e-05, "elapsed_seconds": 172862.07756865025, "projected_total_hours": 75.38707271743914, "projected_total_cost_usd": 104.78803107724039 }, { "iteration": 20500, "train_loss": 3.64292573928833, "validation_loss": 3.791306257247925, "learning_rate": 8.658814581033465e-05, "elapsed_seconds": 177173.67764238268, "projected_total_hours": 75.3828384548891, "projected_total_cost_usd": 104.78214545229584 }, { "iteration": 21000, "train_loss": 3.630474328994751, "validation_loss": 3.7852611541748047, "learning_rate": 8.158823794706145e-05, "elapsed_seconds": 181491.15980650485, "projected_total_hours": 75.38124891434197, "projected_total_cost_usd": 104.77993599093533 }, { "iteration": 21500, "train_loss": 3.621816396713257, "validation_loss": 3.7775344848632812, "learning_rate": 7.673093346653014e-05, "elapsed_seconds": 185801.37050130963, "projected_total_hours": 75.37678338166826, "projected_total_cost_usd": 104.77372890051888 }, { "iteration": 22000, "train_loss": 3.6151795387268066, "validation_loss": 3.7732739448547363, "learning_rate": 7.20286209614895e-05, "elapsed_seconds": 190149.34689292312, "projected_total_hours": 75.38749359138619, "projected_total_cost_usd": 104.7886160920268 }, { "iteration": 22500, "train_loss": 3.606862783432007, "validation_loss": 3.767625093460083, "learning_rate": 6.749329371645251e-05, "elapsed_seconds": 194459.4843392372, "projected_total_hours": 75.3830593611364, "projected_total_cost_usd": 104.7824525119796 }, { "iteration": 23000, "train_loss": 3.598644495010376, "validation_loss": 3.762594699859619, "learning_rate": 6.313651911873158e-05, "elapsed_seconds": 198774.46855475008, "projected_total_hours": 75.38065594950666, "projected_total_cost_usd": 104.77911176981425 }, { "iteration": 23500, "train_loss": 3.5893373489379883, "validation_loss": 3.7594027519226074, "learning_rate": 5.8969409155727336e-05, "elapsed_seconds": 203100.5320912972, "projected_total_hours": 75.3824669936966, "projected_total_cost_usd": 104.78162912123827 }, { "iteration": 24000, "train_loss": 3.583808660507202, "validation_loss": 3.754646062850952, "learning_rate": 5.500259207371864e-05, "elapsed_seconds": 207457.90694285184, "projected_total_hours": 75.3955819213605, "projected_total_cost_usd": 104.7998588706911 }, { "iteration": 24500, "train_loss": 3.57780385017395, "validation_loss": 3.7487640380859375, "learning_rate": 5.1246185270437925e-05, "elapsed_seconds": 211774.69773548096, "projected_total_hours": 75.3937132527676, "projected_total_cost_usd": 104.79726142134695 }, { "iteration": 25000, "train_loss": 3.5740554332733154, "validation_loss": 3.7441320419311523, "learning_rate": 4.7709769490569937e-05, "elapsed_seconds": 216085.9324932173, "projected_total_hours": 75.38998089207803, "projected_total_cost_usd": 104.79207343998846 }, { "iteration": 25500, "train_loss": 3.5674097537994385, "validation_loss": 3.7407143115997314, "learning_rate": 4.440236438998808e-05, "elapsed_seconds": 220397.60610739142, "projected_total_hours": 75.38654500841056, "projected_total_cost_usd": 104.78729756169068 }, { "iteration": 26000, "train_loss": 3.5626490116119385, "validation_loss": 3.7369534969329834, "learning_rate": 4.1332405531052075e-05, "elapsed_seconds": 224747.3975746855, "projected_total_hours": 75.39602867355903, "projected_total_cost_usd": 104.80047985624705 }, { "iteration": 26500, "train_loss": 3.557927131652832, "validation_loss": 3.7329952716827393, "learning_rate": 3.850772286764083e-05, "elapsed_seconds": 229064.31341329962, "projected_total_hours": 75.39433376496444, "projected_total_cost_usd": 104.79812393330057 }, { "iteration": 27000, "train_loss": 3.5516343116760254, "validation_loss": 3.7303378582000732, "learning_rate": 3.593552077479432e-05, "elapsed_seconds": 233381.3956354186, "projected_total_hours": 75.3927553801661, "projected_total_cost_usd": 104.79592997843086 }, { "iteration": 27500, "train_loss": 3.545631170272827, "validation_loss": 3.7279391288757324, "learning_rate": 3.362235967389917e-05, "elapsed_seconds": 237694.1610769406, "projected_total_hours": 75.38986523046398, "projected_total_cost_usd": 104.79191267034493 }, { "iteration": 28000, "train_loss": 3.542006731033325, "validation_loss": 3.7245471477508545, "learning_rate": 3.157413930028304e-05, "elapsed_seconds": 242045.88678893447, "projected_total_hours": 75.39921473385459, "projected_total_cost_usd": 104.80490848005788 }, { "iteration": 28500, "train_loss": 3.5386886596679688, "validation_loss": 3.7228081226348877, "learning_rate": 2.979608365589387e-05, "elapsed_seconds": 246361.35709098727, "projected_total_hours": 75.39714047423976, "projected_total_cost_usd": 104.80202525919326 }, { "iteration": 29000, "train_loss": 3.536064386367798, "validation_loss": 3.7195823192596436, "learning_rate": 2.8292727685442474e-05, "elapsed_seconds": 250676.92014835775, "projected_total_hours": 75.39516563849074, "projected_total_cost_usd": 104.79928023750212 }, { "iteration": 29500, "train_loss": 3.5316381454467773, "validation_loss": 3.7183539867401123, "learning_rate": 2.7067905709990822e-05, "elapsed_seconds": 254991.04394229501, "projected_total_hours": 75.3928322013942, "projected_total_cost_usd": 104.79603675993793 }, { "iteration": 30000, "train_loss": 3.5296361446380615, "validation_loss": 3.716928720474243, "learning_rate": 2.612474164748647e-05, "elapsed_seconds": 259309.55208150297, "projected_total_hours": 75.39185125332585, "projected_total_cost_usd": 104.79467324212293 }, { "iteration": 30500, "train_loss": 3.526557207107544, "validation_loss": 3.71480131149292, "learning_rate": 2.5465641045185573e-05, "elapsed_seconds": 263647.62443690747, "projected_total_hours": 75.39649733441617, "projected_total_cost_usd": 104.80113129483847 }, { "iteration": 31000, "train_loss": 3.5243618488311768, "validation_loss": 3.712334632873535, "learning_rate": 2.5092284944285984e-05, "elapsed_seconds": 267962.22376234084, "projected_total_hours": 75.39438912309589, "projected_total_cost_usd": 104.79820088110328 }, { "iteration": 31400, "train_loss": 3.52117919921875, "validation_loss": 3.712394952774048, "learning_rate": 2.500000057398626e-05, "elapsed_seconds": 271423.6443994045, "projected_total_hours": 75.39545677761237, "projected_total_cost_usd": 104.79968492088118 } ], "environment": { "python": "3.11.10", "platform": "Linux-6.5.0-45-generic-x86_64-with-glibc2.35", "torch": "2.5.1+cu124", "cuda": "12.4", "gpu": "NVIDIA A100 80GB PCIe" }, "data_manifest_sha256": "c05b79cf5753e25fe9276f092a64c5079c5a206fd790d96841477e8cca51530f", "peak_vram_bytes": 43817826304, "gpu_total_bytes": 84974239744, "estimated_training_cost_usd": 104.79968896043354 }