anantoj's picture
Initial commit
8e211ba
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"global_step": 44157,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 1.9774441198450984e-05,
"loss": 2.5305,
"step": 500
},
{
"epoch": 0.07,
"learning_rate": 1.9548429467581584e-05,
"loss": 2.4174,
"step": 1000
},
{
"epoch": 0.1,
"learning_rate": 1.9322417736712188e-05,
"loss": 2.4269,
"step": 1500
},
{
"epoch": 0.14,
"learning_rate": 1.909595307652241e-05,
"loss": 2.3693,
"step": 2000
},
{
"epoch": 0.17,
"learning_rate": 1.8869488416332635e-05,
"loss": 2.3566,
"step": 2500
},
{
"epoch": 0.2,
"learning_rate": 1.8643023756142857e-05,
"loss": 2.3586,
"step": 3000
},
{
"epoch": 0.24,
"learning_rate": 1.841655909595308e-05,
"loss": 2.3499,
"step": 3500
},
{
"epoch": 0.27,
"learning_rate": 1.81900944357633e-05,
"loss": 2.3594,
"step": 4000
},
{
"epoch": 0.31,
"learning_rate": 1.7963629775573522e-05,
"loss": 2.345,
"step": 4500
},
{
"epoch": 0.34,
"learning_rate": 1.7737165115383744e-05,
"loss": 2.321,
"step": 5000
},
{
"epoch": 0.37,
"learning_rate": 1.7510700455193966e-05,
"loss": 2.3003,
"step": 5500
},
{
"epoch": 0.41,
"learning_rate": 1.728423579500419e-05,
"loss": 2.3128,
"step": 6000
},
{
"epoch": 0.44,
"learning_rate": 1.7058224064134795e-05,
"loss": 2.3011,
"step": 6500
},
{
"epoch": 0.48,
"learning_rate": 1.6831759403945017e-05,
"loss": 2.3096,
"step": 7000
},
{
"epoch": 0.51,
"learning_rate": 1.660529474375524e-05,
"loss": 2.3256,
"step": 7500
},
{
"epoch": 0.54,
"learning_rate": 1.637883008356546e-05,
"loss": 2.2922,
"step": 8000
},
{
"epoch": 0.58,
"learning_rate": 1.6152365423375686e-05,
"loss": 2.3021,
"step": 8500
},
{
"epoch": 0.61,
"learning_rate": 1.5925900763185907e-05,
"loss": 2.3011,
"step": 9000
},
{
"epoch": 0.65,
"learning_rate": 1.569943610299613e-05,
"loss": 2.2814,
"step": 9500
},
{
"epoch": 0.68,
"learning_rate": 1.547297144280635e-05,
"loss": 2.2816,
"step": 10000
},
{
"epoch": 0.71,
"learning_rate": 1.5246959711936953e-05,
"loss": 2.2801,
"step": 10500
},
{
"epoch": 0.75,
"learning_rate": 1.5020947981067555e-05,
"loss": 2.2993,
"step": 11000
},
{
"epoch": 0.78,
"learning_rate": 1.4794483320877777e-05,
"loss": 2.3057,
"step": 11500
},
{
"epoch": 0.82,
"learning_rate": 1.4568018660688002e-05,
"loss": 2.2701,
"step": 12000
},
{
"epoch": 0.85,
"learning_rate": 1.4341554000498224e-05,
"loss": 2.2742,
"step": 12500
},
{
"epoch": 0.88,
"learning_rate": 1.4115089340308446e-05,
"loss": 2.2872,
"step": 13000
},
{
"epoch": 0.92,
"learning_rate": 1.388862468011867e-05,
"loss": 2.256,
"step": 13500
},
{
"epoch": 0.95,
"learning_rate": 1.3662160019928891e-05,
"loss": 2.2959,
"step": 14000
},
{
"epoch": 0.99,
"learning_rate": 1.3435695359739113e-05,
"loss": 2.2583,
"step": 14500
},
{
"epoch": 1.0,
"eval_loss": 2.116374969482422,
"eval_runtime": 353.9363,
"eval_samples_per_second": 166.341,
"eval_steps_per_second": 10.397,
"step": 14719
},
{
"epoch": 1.02,
"learning_rate": 1.3209230699549337e-05,
"loss": 2.2496,
"step": 15000
},
{
"epoch": 1.05,
"learning_rate": 1.2983218968679939e-05,
"loss": 2.2561,
"step": 15500
},
{
"epoch": 1.09,
"learning_rate": 1.275675430849016e-05,
"loss": 2.2408,
"step": 16000
},
{
"epoch": 1.12,
"learning_rate": 1.2530289648300384e-05,
"loss": 2.2835,
"step": 16500
},
{
"epoch": 1.15,
"learning_rate": 1.2303824988110606e-05,
"loss": 2.2636,
"step": 17000
},
{
"epoch": 1.19,
"learning_rate": 1.2077360327920828e-05,
"loss": 2.2751,
"step": 17500
},
{
"epoch": 1.22,
"learning_rate": 1.1850895667731051e-05,
"loss": 2.279,
"step": 18000
},
{
"epoch": 1.26,
"learning_rate": 1.1624431007541275e-05,
"loss": 2.27,
"step": 18500
},
{
"epoch": 1.29,
"learning_rate": 1.1398419276671877e-05,
"loss": 2.2436,
"step": 19000
},
{
"epoch": 1.32,
"learning_rate": 1.1171954616482099e-05,
"loss": 2.2575,
"step": 19500
},
{
"epoch": 1.36,
"learning_rate": 1.094548995629232e-05,
"loss": 2.2669,
"step": 20000
},
{
"epoch": 1.39,
"learning_rate": 1.0719025296102544e-05,
"loss": 2.269,
"step": 20500
},
{
"epoch": 1.43,
"learning_rate": 1.0492560635912768e-05,
"loss": 2.2429,
"step": 21000
},
{
"epoch": 1.46,
"learning_rate": 1.026609597572299e-05,
"loss": 2.2424,
"step": 21500
},
{
"epoch": 1.49,
"learning_rate": 1.0039631315533211e-05,
"loss": 2.2549,
"step": 22000
},
{
"epoch": 1.53,
"learning_rate": 9.813166655343435e-06,
"loss": 2.2623,
"step": 22500
},
{
"epoch": 1.56,
"learning_rate": 9.587154924474037e-06,
"loss": 2.2634,
"step": 23000
},
{
"epoch": 1.6,
"learning_rate": 9.360690264284259e-06,
"loss": 2.2674,
"step": 23500
},
{
"epoch": 1.63,
"learning_rate": 9.134225604094482e-06,
"loss": 2.2611,
"step": 24000
},
{
"epoch": 1.66,
"learning_rate": 8.907760943904704e-06,
"loss": 2.2418,
"step": 24500
},
{
"epoch": 1.7,
"learning_rate": 8.681749213035306e-06,
"loss": 2.2493,
"step": 25000
},
{
"epoch": 1.73,
"learning_rate": 8.45528455284553e-06,
"loss": 2.2259,
"step": 25500
},
{
"epoch": 1.77,
"learning_rate": 8.229272821976131e-06,
"loss": 2.2296,
"step": 26000
},
{
"epoch": 1.8,
"learning_rate": 8.002808161786353e-06,
"loss": 2.2258,
"step": 26500
},
{
"epoch": 1.83,
"learning_rate": 7.776343501596577e-06,
"loss": 2.2407,
"step": 27000
},
{
"epoch": 1.87,
"learning_rate": 7.5498788414067996e-06,
"loss": 2.2242,
"step": 27500
},
{
"epoch": 1.9,
"learning_rate": 7.323414181217021e-06,
"loss": 2.2227,
"step": 28000
},
{
"epoch": 1.94,
"learning_rate": 7.096949521027244e-06,
"loss": 2.2264,
"step": 28500
},
{
"epoch": 1.97,
"learning_rate": 6.870484860837466e-06,
"loss": 2.2649,
"step": 29000
},
{
"epoch": 2.0,
"eval_loss": 2.092484951019287,
"eval_runtime": 353.9215,
"eval_samples_per_second": 166.348,
"eval_steps_per_second": 10.398,
"step": 29438
},
{
"epoch": 2.0,
"learning_rate": 6.6440202006476895e-06,
"loss": 2.2415,
"step": 29500
},
{
"epoch": 2.04,
"learning_rate": 6.417555540457911e-06,
"loss": 2.2354,
"step": 30000
},
{
"epoch": 2.07,
"learning_rate": 6.191090880268135e-06,
"loss": 2.2296,
"step": 30500
},
{
"epoch": 2.11,
"learning_rate": 5.965079149398737e-06,
"loss": 2.241,
"step": 31000
},
{
"epoch": 2.14,
"learning_rate": 5.738614489208959e-06,
"loss": 2.2279,
"step": 31500
},
{
"epoch": 2.17,
"learning_rate": 5.512149829019182e-06,
"loss": 2.2405,
"step": 32000
},
{
"epoch": 2.21,
"learning_rate": 5.285685168829404e-06,
"loss": 2.2379,
"step": 32500
},
{
"epoch": 2.24,
"learning_rate": 5.059673437960007e-06,
"loss": 2.2386,
"step": 33000
},
{
"epoch": 2.28,
"learning_rate": 4.833208777770229e-06,
"loss": 2.2298,
"step": 33500
},
{
"epoch": 2.31,
"learning_rate": 4.6067441175804515e-06,
"loss": 2.2259,
"step": 34000
},
{
"epoch": 2.34,
"learning_rate": 4.380279457390674e-06,
"loss": 2.2537,
"step": 34500
},
{
"epoch": 2.38,
"learning_rate": 4.153814797200897e-06,
"loss": 2.2521,
"step": 35000
},
{
"epoch": 2.41,
"learning_rate": 3.927803066331499e-06,
"loss": 2.2049,
"step": 35500
},
{
"epoch": 2.45,
"learning_rate": 3.701338406141722e-06,
"loss": 2.2317,
"step": 36000
},
{
"epoch": 2.48,
"learning_rate": 3.4748737459519442e-06,
"loss": 2.2476,
"step": 36500
},
{
"epoch": 2.51,
"learning_rate": 3.248409085762167e-06,
"loss": 2.2435,
"step": 37000
},
{
"epoch": 2.55,
"learning_rate": 3.0219444255723896e-06,
"loss": 2.2523,
"step": 37500
},
{
"epoch": 2.58,
"learning_rate": 2.795479765382612e-06,
"loss": 2.2264,
"step": 38000
},
{
"epoch": 2.62,
"learning_rate": 2.5694680345132143e-06,
"loss": 2.2287,
"step": 38500
},
{
"epoch": 2.65,
"learning_rate": 2.343003374323437e-06,
"loss": 2.2238,
"step": 39000
},
{
"epoch": 2.68,
"learning_rate": 2.116991643454039e-06,
"loss": 2.2383,
"step": 39500
},
{
"epoch": 2.72,
"learning_rate": 1.8905269832642617e-06,
"loss": 2.234,
"step": 40000
},
{
"epoch": 2.75,
"learning_rate": 1.6640623230744844e-06,
"loss": 2.2298,
"step": 40500
},
{
"epoch": 2.79,
"learning_rate": 1.4375976628847069e-06,
"loss": 2.2237,
"step": 41000
},
{
"epoch": 2.82,
"learning_rate": 1.2111330026949296e-06,
"loss": 2.2489,
"step": 41500
},
{
"epoch": 2.85,
"learning_rate": 9.846683425051522e-07,
"loss": 2.2263,
"step": 42000
},
{
"epoch": 2.89,
"learning_rate": 7.582036823153747e-07,
"loss": 2.2223,
"step": 42500
},
{
"epoch": 2.92,
"learning_rate": 5.317390221255973e-07,
"loss": 2.2229,
"step": 43000
},
{
"epoch": 2.96,
"learning_rate": 3.0527436193581997e-07,
"loss": 2.2121,
"step": 43500
},
{
"epoch": 2.99,
"learning_rate": 7.880970174604252e-08,
"loss": 2.209,
"step": 44000
},
{
"epoch": 3.0,
"eval_loss": 2.086761713027954,
"eval_runtime": 355.3374,
"eval_samples_per_second": 165.685,
"eval_steps_per_second": 10.356,
"step": 44157
}
],
"max_steps": 44157,
"num_train_epochs": 3,
"total_flos": 6.916093810478285e+16,
"trial_name": null,
"trial_params": null
}