1084 lines
28 KiB
JSON
1084 lines
28 KiB
JSON
|
{
|
||
|
|
"best_metric": 1.805204153060913,
|
||
|
|
"best_model_checkpoint": "/home/sr2464/scratch/C2S_Files/C2S_training_runs/multicell_v2_pretraining_runs/finetuning-EleutherAI/pythia-410m-multicell_v2_pretraining-2024-07-28_14-10-44/checkpoint-7000",
|
||
|
|
"epoch": 0.024296135179068302,
|
||
|
|
"eval_steps": 100,
|
||
|
|
"global_step": 7000,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.00034708764541526143,
|
||
|
|
"grad_norm": 12.638826370239258,
|
||
|
|
"learning_rate": 3.470776065528252e-07,
|
||
|
|
"loss": 4.1006,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00034708764541526143,
|
||
|
|
"eval_loss": 3.750605821609497,
|
||
|
|
"eval_runtime": 28.5808,
|
||
|
|
"eval_samples_per_second": 16.69,
|
||
|
|
"eval_steps_per_second": 0.35,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0006941752908305229,
|
||
|
|
"grad_norm": 7.417098522186279,
|
||
|
|
"learning_rate": 6.941552131056504e-07,
|
||
|
|
"loss": 3.4941,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0006941752908305229,
|
||
|
|
"eval_loss": 3.152513265609741,
|
||
|
|
"eval_runtime": 27.3056,
|
||
|
|
"eval_samples_per_second": 17.469,
|
||
|
|
"eval_steps_per_second": 0.366,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0010412629362457843,
|
||
|
|
"grad_norm": 5.065331935882568,
|
||
|
|
"learning_rate": 1.0412328196584756e-06,
|
||
|
|
"loss": 3.0197,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0010412629362457843,
|
||
|
|
"eval_loss": 2.8438754081726074,
|
||
|
|
"eval_runtime": 27.3207,
|
||
|
|
"eval_samples_per_second": 17.459,
|
||
|
|
"eval_steps_per_second": 0.366,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0013883505816610457,
|
||
|
|
"grad_norm": 5.113323211669922,
|
||
|
|
"learning_rate": 1.3883104262113009e-06,
|
||
|
|
"loss": 2.7915,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0013883505816610457,
|
||
|
|
"eval_loss": 2.6905336380004883,
|
||
|
|
"eval_runtime": 27.6367,
|
||
|
|
"eval_samples_per_second": 17.26,
|
||
|
|
"eval_steps_per_second": 0.362,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0017354382270763071,
|
||
|
|
"grad_norm": 6.473913192749023,
|
||
|
|
"learning_rate": 1.7353880327641261e-06,
|
||
|
|
"loss": 2.6697,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0017354382270763071,
|
||
|
|
"eval_loss": 2.609987735748291,
|
||
|
|
"eval_runtime": 27.5233,
|
||
|
|
"eval_samples_per_second": 17.331,
|
||
|
|
"eval_steps_per_second": 0.363,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0020825258724915686,
|
||
|
|
"grad_norm": 6.615407466888428,
|
||
|
|
"learning_rate": 2.082465639316951e-06,
|
||
|
|
"loss": 2.5959,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0020825258724915686,
|
||
|
|
"eval_loss": 2.5381109714508057,
|
||
|
|
"eval_runtime": 27.411,
|
||
|
|
"eval_samples_per_second": 17.402,
|
||
|
|
"eval_steps_per_second": 0.365,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00242961351790683,
|
||
|
|
"grad_norm": 8.234296798706055,
|
||
|
|
"learning_rate": 2.429543245869777e-06,
|
||
|
|
"loss": 2.5239,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00242961351790683,
|
||
|
|
"eval_loss": 2.4786229133605957,
|
||
|
|
"eval_runtime": 27.9876,
|
||
|
|
"eval_samples_per_second": 17.043,
|
||
|
|
"eval_steps_per_second": 0.357,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0027767011633220914,
|
||
|
|
"grad_norm": 3.987603187561035,
|
||
|
|
"learning_rate": 2.7766208524226017e-06,
|
||
|
|
"loss": 2.4544,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0027767011633220914,
|
||
|
|
"eval_loss": 2.4083495140075684,
|
||
|
|
"eval_runtime": 27.5819,
|
||
|
|
"eval_samples_per_second": 17.294,
|
||
|
|
"eval_steps_per_second": 0.363,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.003123788808737353,
|
||
|
|
"grad_norm": 3.854851484298706,
|
||
|
|
"learning_rate": 3.123698458975427e-06,
|
||
|
|
"loss": 2.3861,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.003123788808737353,
|
||
|
|
"eval_loss": 2.336987257003784,
|
||
|
|
"eval_runtime": 27.6168,
|
||
|
|
"eval_samples_per_second": 17.272,
|
||
|
|
"eval_steps_per_second": 0.362,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0034708764541526143,
|
||
|
|
"grad_norm": 3.7287888526916504,
|
||
|
|
"learning_rate": 3.4707760655282523e-06,
|
||
|
|
"loss": 2.3199,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0034708764541526143,
|
||
|
|
"eval_loss": 2.2719526290893555,
|
||
|
|
"eval_runtime": 26.9345,
|
||
|
|
"eval_samples_per_second": 17.71,
|
||
|
|
"eval_steps_per_second": 0.371,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0038179640995678757,
|
||
|
|
"grad_norm": 3.5708882808685303,
|
||
|
|
"learning_rate": 3.817853672081077e-06,
|
||
|
|
"loss": 2.2467,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0038179640995678757,
|
||
|
|
"eval_loss": 2.206721067428589,
|
||
|
|
"eval_runtime": 27.0387,
|
||
|
|
"eval_samples_per_second": 17.641,
|
||
|
|
"eval_steps_per_second": 0.37,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.004165051744983137,
|
||
|
|
"grad_norm": 3.284808397293091,
|
||
|
|
"learning_rate": 4.164931278633902e-06,
|
||
|
|
"loss": 2.2017,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.004165051744983137,
|
||
|
|
"eval_loss": 2.1687662601470947,
|
||
|
|
"eval_runtime": 26.9942,
|
||
|
|
"eval_samples_per_second": 17.67,
|
||
|
|
"eval_steps_per_second": 0.37,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.004512139390398399,
|
||
|
|
"grad_norm": 3.306734323501587,
|
||
|
|
"learning_rate": 4.5120088851867285e-06,
|
||
|
|
"loss": 2.1612,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.004512139390398399,
|
||
|
|
"eval_loss": 2.129286766052246,
|
||
|
|
"eval_runtime": 27.1243,
|
||
|
|
"eval_samples_per_second": 17.586,
|
||
|
|
"eval_steps_per_second": 0.369,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00485922703581366,
|
||
|
|
"grad_norm": 4.48453950881958,
|
||
|
|
"learning_rate": 4.859086491739554e-06,
|
||
|
|
"loss": 2.1302,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00485922703581366,
|
||
|
|
"eval_loss": 2.1066970825195312,
|
||
|
|
"eval_runtime": 26.7667,
|
||
|
|
"eval_samples_per_second": 17.821,
|
||
|
|
"eval_steps_per_second": 0.374,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0052063146812289214,
|
||
|
|
"grad_norm": 3.936067819595337,
|
||
|
|
"learning_rate": 5.206164098292378e-06,
|
||
|
|
"loss": 2.1019,
|
||
|
|
"step": 1500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0052063146812289214,
|
||
|
|
"eval_loss": 2.079268217086792,
|
||
|
|
"eval_runtime": 26.9038,
|
||
|
|
"eval_samples_per_second": 17.73,
|
||
|
|
"eval_steps_per_second": 0.372,
|
||
|
|
"step": 1500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.005553402326644183,
|
||
|
|
"grad_norm": 3.049617290496826,
|
||
|
|
"learning_rate": 5.5532417048452035e-06,
|
||
|
|
"loss": 2.0839,
|
||
|
|
"step": 1600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.005553402326644183,
|
||
|
|
"eval_loss": 2.063786268234253,
|
||
|
|
"eval_runtime": 26.8676,
|
||
|
|
"eval_samples_per_second": 17.754,
|
||
|
|
"eval_steps_per_second": 0.372,
|
||
|
|
"step": 1600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.005900489972059444,
|
||
|
|
"grad_norm": 2.732224464416504,
|
||
|
|
"learning_rate": 5.900319311398029e-06,
|
||
|
|
"loss": 2.0707,
|
||
|
|
"step": 1700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.005900489972059444,
|
||
|
|
"eval_loss": 2.046088695526123,
|
||
|
|
"eval_runtime": 26.9915,
|
||
|
|
"eval_samples_per_second": 17.672,
|
||
|
|
"eval_steps_per_second": 0.37,
|
||
|
|
"step": 1700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.006247577617474706,
|
||
|
|
"grad_norm": 2.1865687370300293,
|
||
|
|
"learning_rate": 6.247396917950854e-06,
|
||
|
|
"loss": 2.0527,
|
||
|
|
"step": 1800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.006247577617474706,
|
||
|
|
"eval_loss": 2.031470537185669,
|
||
|
|
"eval_runtime": 26.9114,
|
||
|
|
"eval_samples_per_second": 17.725,
|
||
|
|
"eval_steps_per_second": 0.372,
|
||
|
|
"step": 1800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.006594665262889967,
|
||
|
|
"grad_norm": 2.854724407196045,
|
||
|
|
"learning_rate": 6.59447452450368e-06,
|
||
|
|
"loss": 2.0407,
|
||
|
|
"step": 1900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.006594665262889967,
|
||
|
|
"eval_loss": 2.0221400260925293,
|
||
|
|
"eval_runtime": 26.7594,
|
||
|
|
"eval_samples_per_second": 17.825,
|
||
|
|
"eval_steps_per_second": 0.374,
|
||
|
|
"step": 1900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.006941752908305229,
|
||
|
|
"grad_norm": 2.716592788696289,
|
||
|
|
"learning_rate": 6.9415521310565046e-06,
|
||
|
|
"loss": 2.0369,
|
||
|
|
"step": 2000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.006941752908305229,
|
||
|
|
"eval_loss": 2.0095324516296387,
|
||
|
|
"eval_runtime": 26.7666,
|
||
|
|
"eval_samples_per_second": 17.821,
|
||
|
|
"eval_steps_per_second": 0.374,
|
||
|
|
"step": 2000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00728884055372049,
|
||
|
|
"grad_norm": 3.099235773086548,
|
||
|
|
"learning_rate": 7.28862973760933e-06,
|
||
|
|
"loss": 2.0172,
|
||
|
|
"step": 2100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00728884055372049,
|
||
|
|
"eval_loss": 1.9926340579986572,
|
||
|
|
"eval_runtime": 26.7511,
|
||
|
|
"eval_samples_per_second": 17.831,
|
||
|
|
"eval_steps_per_second": 0.374,
|
||
|
|
"step": 2100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0076359281991357515,
|
||
|
|
"grad_norm": 2.7901623249053955,
|
||
|
|
"learning_rate": 7.635707344162154e-06,
|
||
|
|
"loss": 2.0011,
|
||
|
|
"step": 2200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0076359281991357515,
|
||
|
|
"eval_loss": 1.9787664413452148,
|
||
|
|
"eval_runtime": 26.8686,
|
||
|
|
"eval_samples_per_second": 17.753,
|
||
|
|
"eval_steps_per_second": 0.372,
|
||
|
|
"step": 2200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.007983015844551014,
|
||
|
|
"grad_norm": 1.9205849170684814,
|
||
|
|
"learning_rate": 7.982784950714981e-06,
|
||
|
|
"loss": 1.9848,
|
||
|
|
"step": 2300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.007983015844551014,
|
||
|
|
"eval_loss": 1.9650758504867554,
|
||
|
|
"eval_runtime": 26.6567,
|
||
|
|
"eval_samples_per_second": 17.894,
|
||
|
|
"eval_steps_per_second": 0.375,
|
||
|
|
"step": 2300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.008330103489966274,
|
||
|
|
"grad_norm": 2.2833688259124756,
|
||
|
|
"learning_rate": 8.329862557267805e-06,
|
||
|
|
"loss": 1.9804,
|
||
|
|
"step": 2400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.008330103489966274,
|
||
|
|
"eval_loss": 1.95713472366333,
|
||
|
|
"eval_runtime": 26.6415,
|
||
|
|
"eval_samples_per_second": 17.904,
|
||
|
|
"eval_steps_per_second": 0.375,
|
||
|
|
"step": 2400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.008677191135381537,
|
||
|
|
"grad_norm": 2.6494622230529785,
|
||
|
|
"learning_rate": 8.67694016382063e-06,
|
||
|
|
"loss": 1.9622,
|
||
|
|
"step": 2500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.008677191135381537,
|
||
|
|
"eval_loss": 1.949569821357727,
|
||
|
|
"eval_runtime": 26.7134,
|
||
|
|
"eval_samples_per_second": 17.856,
|
||
|
|
"eval_steps_per_second": 0.374,
|
||
|
|
"step": 2500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.009024278780796797,
|
||
|
|
"grad_norm": 1.9379554986953735,
|
||
|
|
"learning_rate": 9.024017770373457e-06,
|
||
|
|
"loss": 1.9624,
|
||
|
|
"step": 2600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.009024278780796797,
|
||
|
|
"eval_loss": 1.9367306232452393,
|
||
|
|
"eval_runtime": 26.6328,
|
||
|
|
"eval_samples_per_second": 17.91,
|
||
|
|
"eval_steps_per_second": 0.375,
|
||
|
|
"step": 2600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00937136642621206,
|
||
|
|
"grad_norm": 1.797877311706543,
|
||
|
|
"learning_rate": 9.37109537692628e-06,
|
||
|
|
"loss": 1.9466,
|
||
|
|
"step": 2700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00937136642621206,
|
||
|
|
"eval_loss": 1.9342317581176758,
|
||
|
|
"eval_runtime": 26.6027,
|
||
|
|
"eval_samples_per_second": 17.931,
|
||
|
|
"eval_steps_per_second": 0.376,
|
||
|
|
"step": 2700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00971845407162732,
|
||
|
|
"grad_norm": 2.123497724533081,
|
||
|
|
"learning_rate": 9.718172983479108e-06,
|
||
|
|
"loss": 1.9421,
|
||
|
|
"step": 2800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.00971845407162732,
|
||
|
|
"eval_loss": 1.9269955158233643,
|
||
|
|
"eval_runtime": 26.3282,
|
||
|
|
"eval_samples_per_second": 18.117,
|
||
|
|
"eval_steps_per_second": 0.38,
|
||
|
|
"step": 2800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010065541717042582,
|
||
|
|
"grad_norm": 2.386960029602051,
|
||
|
|
"learning_rate": 1.0065250590031931e-05,
|
||
|
|
"loss": 1.9395,
|
||
|
|
"step": 2900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010065541717042582,
|
||
|
|
"eval_loss": 1.9186092615127563,
|
||
|
|
"eval_runtime": 26.5488,
|
||
|
|
"eval_samples_per_second": 17.967,
|
||
|
|
"eval_steps_per_second": 0.377,
|
||
|
|
"step": 2900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010412629362457843,
|
||
|
|
"grad_norm": 1.994126558303833,
|
||
|
|
"learning_rate": 1.0412328196584756e-05,
|
||
|
|
"loss": 1.9242,
|
||
|
|
"step": 3000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010412629362457843,
|
||
|
|
"eval_loss": 1.9123972654342651,
|
||
|
|
"eval_runtime": 26.4618,
|
||
|
|
"eval_samples_per_second": 18.026,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 3000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010759717007873105,
|
||
|
|
"grad_norm": 1.7175105810165405,
|
||
|
|
"learning_rate": 1.0759405803137582e-05,
|
||
|
|
"loss": 1.9232,
|
||
|
|
"step": 3100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010759717007873105,
|
||
|
|
"eval_loss": 1.903687834739685,
|
||
|
|
"eval_runtime": 26.5449,
|
||
|
|
"eval_samples_per_second": 17.97,
|
||
|
|
"eval_steps_per_second": 0.377,
|
||
|
|
"step": 3100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.011106804653288366,
|
||
|
|
"grad_norm": 1.8057008981704712,
|
||
|
|
"learning_rate": 1.1106483409690407e-05,
|
||
|
|
"loss": 1.923,
|
||
|
|
"step": 3200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.011106804653288366,
|
||
|
|
"eval_loss": 1.9004727602005005,
|
||
|
|
"eval_runtime": 26.3249,
|
||
|
|
"eval_samples_per_second": 18.12,
|
||
|
|
"eval_steps_per_second": 0.38,
|
||
|
|
"step": 3200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.011453892298703628,
|
||
|
|
"grad_norm": 1.6960588693618774,
|
||
|
|
"learning_rate": 1.1453561016243232e-05,
|
||
|
|
"loss": 1.9102,
|
||
|
|
"step": 3300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.011453892298703628,
|
||
|
|
"eval_loss": 1.8942128419876099,
|
||
|
|
"eval_runtime": 26.351,
|
||
|
|
"eval_samples_per_second": 18.102,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 3300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.011800979944118889,
|
||
|
|
"grad_norm": 1.8168739080429077,
|
||
|
|
"learning_rate": 1.1800638622796058e-05,
|
||
|
|
"loss": 1.9074,
|
||
|
|
"step": 3400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.011800979944118889,
|
||
|
|
"eval_loss": 1.889912486076355,
|
||
|
|
"eval_runtime": 26.273,
|
||
|
|
"eval_samples_per_second": 18.155,
|
||
|
|
"eval_steps_per_second": 0.381,
|
||
|
|
"step": 3400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.012148067589534151,
|
||
|
|
"grad_norm": 1.8745700120925903,
|
||
|
|
"learning_rate": 1.2147716229348883e-05,
|
||
|
|
"loss": 1.9053,
|
||
|
|
"step": 3500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.012148067589534151,
|
||
|
|
"eval_loss": 1.8899447917938232,
|
||
|
|
"eval_runtime": 26.3575,
|
||
|
|
"eval_samples_per_second": 18.097,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 3500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.012495155234949411,
|
||
|
|
"grad_norm": 1.77675199508667,
|
||
|
|
"learning_rate": 1.2494793835901708e-05,
|
||
|
|
"loss": 1.8983,
|
||
|
|
"step": 3600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.012495155234949411,
|
||
|
|
"eval_loss": 1.8839936256408691,
|
||
|
|
"eval_runtime": 26.4506,
|
||
|
|
"eval_samples_per_second": 18.034,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 3600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.012842242880364674,
|
||
|
|
"grad_norm": 1.473862648010254,
|
||
|
|
"learning_rate": 1.2841871442454533e-05,
|
||
|
|
"loss": 1.8998,
|
||
|
|
"step": 3700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.012842242880364674,
|
||
|
|
"eval_loss": 1.8791130781173706,
|
||
|
|
"eval_runtime": 26.7984,
|
||
|
|
"eval_samples_per_second": 17.8,
|
||
|
|
"eval_steps_per_second": 0.373,
|
||
|
|
"step": 3700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013189330525779934,
|
||
|
|
"grad_norm": 1.486438512802124,
|
||
|
|
"learning_rate": 1.318894904900736e-05,
|
||
|
|
"loss": 1.8951,
|
||
|
|
"step": 3800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013189330525779934,
|
||
|
|
"eval_loss": 1.8784898519515991,
|
||
|
|
"eval_runtime": 26.3412,
|
||
|
|
"eval_samples_per_second": 18.109,
|
||
|
|
"eval_steps_per_second": 0.38,
|
||
|
|
"step": 3800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013536418171195197,
|
||
|
|
"grad_norm": 1.2373360395431519,
|
||
|
|
"learning_rate": 1.3536026655560182e-05,
|
||
|
|
"loss": 1.887,
|
||
|
|
"step": 3900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013536418171195197,
|
||
|
|
"eval_loss": 1.870314121246338,
|
||
|
|
"eval_runtime": 26.436,
|
||
|
|
"eval_samples_per_second": 18.044,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 3900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013883505816610457,
|
||
|
|
"grad_norm": 1.5347727537155151,
|
||
|
|
"learning_rate": 1.3883104262113009e-05,
|
||
|
|
"loss": 1.8951,
|
||
|
|
"step": 4000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013883505816610457,
|
||
|
|
"eval_loss": 1.8662590980529785,
|
||
|
|
"eval_runtime": 26.4553,
|
||
|
|
"eval_samples_per_second": 18.03,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 4000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01423059346202572,
|
||
|
|
"grad_norm": 1.771752953529358,
|
||
|
|
"learning_rate": 1.4230181868665834e-05,
|
||
|
|
"loss": 1.8854,
|
||
|
|
"step": 4100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01423059346202572,
|
||
|
|
"eval_loss": 1.8646233081817627,
|
||
|
|
"eval_runtime": 26.398,
|
||
|
|
"eval_samples_per_second": 18.07,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 4100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01457768110744098,
|
||
|
|
"grad_norm": 2.278961420059204,
|
||
|
|
"learning_rate": 1.457725947521866e-05,
|
||
|
|
"loss": 1.8829,
|
||
|
|
"step": 4200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01457768110744098,
|
||
|
|
"eval_loss": 1.8655200004577637,
|
||
|
|
"eval_runtime": 26.4296,
|
||
|
|
"eval_samples_per_second": 18.048,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 4200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.014924768752856242,
|
||
|
|
"grad_norm": 1.965826392173767,
|
||
|
|
"learning_rate": 1.4924337081771487e-05,
|
||
|
|
"loss": 1.8768,
|
||
|
|
"step": 4300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.014924768752856242,
|
||
|
|
"eval_loss": 1.861385464668274,
|
||
|
|
"eval_runtime": 26.4002,
|
||
|
|
"eval_samples_per_second": 18.068,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 4300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015271856398271503,
|
||
|
|
"grad_norm": 1.2046449184417725,
|
||
|
|
"learning_rate": 1.527141468832431e-05,
|
||
|
|
"loss": 1.8789,
|
||
|
|
"step": 4400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015271856398271503,
|
||
|
|
"eval_loss": 1.8564562797546387,
|
||
|
|
"eval_runtime": 26.5199,
|
||
|
|
"eval_samples_per_second": 17.986,
|
||
|
|
"eval_steps_per_second": 0.377,
|
||
|
|
"step": 4400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015618944043686765,
|
||
|
|
"grad_norm": 1.2664066553115845,
|
||
|
|
"learning_rate": 1.5618492294877134e-05,
|
||
|
|
"loss": 1.8676,
|
||
|
|
"step": 4500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015618944043686765,
|
||
|
|
"eval_loss": 1.8561848402023315,
|
||
|
|
"eval_runtime": 26.3757,
|
||
|
|
"eval_samples_per_second": 18.085,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 4500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015966031689102027,
|
||
|
|
"grad_norm": 1.242021918296814,
|
||
|
|
"learning_rate": 1.5965569901429962e-05,
|
||
|
|
"loss": 1.8747,
|
||
|
|
"step": 4600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.015966031689102027,
|
||
|
|
"eval_loss": 1.8486684560775757,
|
||
|
|
"eval_runtime": 26.3838,
|
||
|
|
"eval_samples_per_second": 18.079,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 4600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.016313119334517286,
|
||
|
|
"grad_norm": 1.7827301025390625,
|
||
|
|
"learning_rate": 1.6312647507982788e-05,
|
||
|
|
"loss": 1.8675,
|
||
|
|
"step": 4700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.016313119334517286,
|
||
|
|
"eval_loss": 1.848850131034851,
|
||
|
|
"eval_runtime": 26.3928,
|
||
|
|
"eval_samples_per_second": 18.073,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 4700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01666020697993255,
|
||
|
|
"grad_norm": 1.4154130220413208,
|
||
|
|
"learning_rate": 1.665972511453561e-05,
|
||
|
|
"loss": 1.8646,
|
||
|
|
"step": 4800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01666020697993255,
|
||
|
|
"eval_loss": 1.8437469005584717,
|
||
|
|
"eval_runtime": 26.3717,
|
||
|
|
"eval_samples_per_second": 18.088,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 4800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01700729462534781,
|
||
|
|
"grad_norm": 1.556569218635559,
|
||
|
|
"learning_rate": 1.7006802721088435e-05,
|
||
|
|
"loss": 1.8633,
|
||
|
|
"step": 4900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01700729462534781,
|
||
|
|
"eval_loss": 1.8424543142318726,
|
||
|
|
"eval_runtime": 26.5267,
|
||
|
|
"eval_samples_per_second": 17.982,
|
||
|
|
"eval_steps_per_second": 0.377,
|
||
|
|
"step": 4900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.017354382270763073,
|
||
|
|
"grad_norm": 1.2393227815628052,
|
||
|
|
"learning_rate": 1.735388032764126e-05,
|
||
|
|
"loss": 1.8581,
|
||
|
|
"step": 5000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.017354382270763073,
|
||
|
|
"eval_loss": 1.8405553102493286,
|
||
|
|
"eval_runtime": 26.3219,
|
||
|
|
"eval_samples_per_second": 18.122,
|
||
|
|
"eval_steps_per_second": 0.38,
|
||
|
|
"step": 5000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.017701469916178332,
|
||
|
|
"grad_norm": 1.141632318496704,
|
||
|
|
"learning_rate": 1.770095793419409e-05,
|
||
|
|
"loss": 1.8589,
|
||
|
|
"step": 5100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.017701469916178332,
|
||
|
|
"eval_loss": 1.8358362913131714,
|
||
|
|
"eval_runtime": 26.3268,
|
||
|
|
"eval_samples_per_second": 18.118,
|
||
|
|
"eval_steps_per_second": 0.38,
|
||
|
|
"step": 5100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.018048557561593594,
|
||
|
|
"grad_norm": 1.1542439460754395,
|
||
|
|
"learning_rate": 1.8048035540746914e-05,
|
||
|
|
"loss": 1.8583,
|
||
|
|
"step": 5200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.018048557561593594,
|
||
|
|
"eval_loss": 1.8365525007247925,
|
||
|
|
"eval_runtime": 26.4883,
|
||
|
|
"eval_samples_per_second": 18.008,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 5200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.018395645207008857,
|
||
|
|
"grad_norm": 1.0082952976226807,
|
||
|
|
"learning_rate": 1.8395113147299736e-05,
|
||
|
|
"loss": 1.8523,
|
||
|
|
"step": 5300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.018395645207008857,
|
||
|
|
"eval_loss": 1.83348548412323,
|
||
|
|
"eval_runtime": 26.7213,
|
||
|
|
"eval_samples_per_second": 17.851,
|
||
|
|
"eval_steps_per_second": 0.374,
|
||
|
|
"step": 5300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01874273285242412,
|
||
|
|
"grad_norm": 1.3992128372192383,
|
||
|
|
"learning_rate": 1.874219075385256e-05,
|
||
|
|
"loss": 1.8557,
|
||
|
|
"step": 5400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01874273285242412,
|
||
|
|
"eval_loss": 1.830024003982544,
|
||
|
|
"eval_runtime": 26.3985,
|
||
|
|
"eval_samples_per_second": 18.069,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 5400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.019089820497839378,
|
||
|
|
"grad_norm": 1.0521454811096191,
|
||
|
|
"learning_rate": 1.9089268360405386e-05,
|
||
|
|
"loss": 1.8541,
|
||
|
|
"step": 5500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.019089820497839378,
|
||
|
|
"eval_loss": 1.8300797939300537,
|
||
|
|
"eval_runtime": 26.4761,
|
||
|
|
"eval_samples_per_second": 18.016,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 5500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01943690814325464,
|
||
|
|
"grad_norm": 1.2668583393096924,
|
||
|
|
"learning_rate": 1.9436345966958215e-05,
|
||
|
|
"loss": 1.8453,
|
||
|
|
"step": 5600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.01943690814325464,
|
||
|
|
"eval_loss": 1.8309643268585205,
|
||
|
|
"eval_runtime": 26.3911,
|
||
|
|
"eval_samples_per_second": 18.074,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 5600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.019783995788669902,
|
||
|
|
"grad_norm": 1.09402334690094,
|
||
|
|
"learning_rate": 1.9783423573511037e-05,
|
||
|
|
"loss": 1.8536,
|
||
|
|
"step": 5700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.019783995788669902,
|
||
|
|
"eval_loss": 1.8289095163345337,
|
||
|
|
"eval_runtime": 26.4192,
|
||
|
|
"eval_samples_per_second": 18.055,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 5700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020131083434085165,
|
||
|
|
"grad_norm": 0.960943341255188,
|
||
|
|
"learning_rate": 2.0130501180063862e-05,
|
||
|
|
"loss": 1.8509,
|
||
|
|
"step": 5800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020131083434085165,
|
||
|
|
"eval_loss": 1.824504017829895,
|
||
|
|
"eval_runtime": 26.471,
|
||
|
|
"eval_samples_per_second": 18.02,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 5800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020478171079500423,
|
||
|
|
"grad_norm": 1.0958678722381592,
|
||
|
|
"learning_rate": 2.0477578786616688e-05,
|
||
|
|
"loss": 1.8452,
|
||
|
|
"step": 5900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020478171079500423,
|
||
|
|
"eval_loss": 1.8239096403121948,
|
||
|
|
"eval_runtime": 26.3508,
|
||
|
|
"eval_samples_per_second": 18.102,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 5900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020825258724915686,
|
||
|
|
"grad_norm": 1.1169177293777466,
|
||
|
|
"learning_rate": 2.0824656393169513e-05,
|
||
|
|
"loss": 1.8431,
|
||
|
|
"step": 6000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.020825258724915686,
|
||
|
|
"eval_loss": 1.8211857080459595,
|
||
|
|
"eval_runtime": 26.3842,
|
||
|
|
"eval_samples_per_second": 18.079,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 6000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.021172346370330948,
|
||
|
|
"grad_norm": 1.2566906213760376,
|
||
|
|
"learning_rate": 2.117173399972234e-05,
|
||
|
|
"loss": 1.8399,
|
||
|
|
"step": 6100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.021172346370330948,
|
||
|
|
"eval_loss": 1.8177697658538818,
|
||
|
|
"eval_runtime": 26.4481,
|
||
|
|
"eval_samples_per_second": 18.035,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 6100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02151943401574621,
|
||
|
|
"grad_norm": 1.0141446590423584,
|
||
|
|
"learning_rate": 2.1518811606275163e-05,
|
||
|
|
"loss": 1.8412,
|
||
|
|
"step": 6200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02151943401574621,
|
||
|
|
"eval_loss": 1.8186702728271484,
|
||
|
|
"eval_runtime": 26.5203,
|
||
|
|
"eval_samples_per_second": 17.986,
|
||
|
|
"eval_steps_per_second": 0.377,
|
||
|
|
"step": 6200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.021866521661161473,
|
||
|
|
"grad_norm": 1.1088286638259888,
|
||
|
|
"learning_rate": 2.186588921282799e-05,
|
||
|
|
"loss": 1.8433,
|
||
|
|
"step": 6300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.021866521661161473,
|
||
|
|
"eval_loss": 1.8170266151428223,
|
||
|
|
"eval_runtime": 26.8479,
|
||
|
|
"eval_samples_per_second": 17.767,
|
||
|
|
"eval_steps_per_second": 0.372,
|
||
|
|
"step": 6300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02221360930657673,
|
||
|
|
"grad_norm": 1.0931763648986816,
|
||
|
|
"learning_rate": 2.2212966819380814e-05,
|
||
|
|
"loss": 1.8363,
|
||
|
|
"step": 6400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02221360930657673,
|
||
|
|
"eval_loss": 1.81370210647583,
|
||
|
|
"eval_runtime": 26.4259,
|
||
|
|
"eval_samples_per_second": 18.05,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 6400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.022560696951991994,
|
||
|
|
"grad_norm": 1.040462851524353,
|
||
|
|
"learning_rate": 2.256004442593364e-05,
|
||
|
|
"loss": 1.8339,
|
||
|
|
"step": 6500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.022560696951991994,
|
||
|
|
"eval_loss": 1.8151589632034302,
|
||
|
|
"eval_runtime": 26.44,
|
||
|
|
"eval_samples_per_second": 18.041,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 6500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.022907784597407256,
|
||
|
|
"grad_norm": 1.0170607566833496,
|
||
|
|
"learning_rate": 2.2907122032486464e-05,
|
||
|
|
"loss": 1.833,
|
||
|
|
"step": 6600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.022907784597407256,
|
||
|
|
"eval_loss": 1.8107798099517822,
|
||
|
|
"eval_runtime": 26.3941,
|
||
|
|
"eval_samples_per_second": 18.072,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 6600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02325487224282252,
|
||
|
|
"grad_norm": 0.813853919506073,
|
||
|
|
"learning_rate": 2.325419963903929e-05,
|
||
|
|
"loss": 1.8338,
|
||
|
|
"step": 6700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02325487224282252,
|
||
|
|
"eval_loss": 1.8114594221115112,
|
||
|
|
"eval_runtime": 26.3741,
|
||
|
|
"eval_samples_per_second": 18.086,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 6700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.023601959888237777,
|
||
|
|
"grad_norm": 1.4288556575775146,
|
||
|
|
"learning_rate": 2.3601277245592115e-05,
|
||
|
|
"loss": 1.8301,
|
||
|
|
"step": 6800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.023601959888237777,
|
||
|
|
"eval_loss": 1.8073900938034058,
|
||
|
|
"eval_runtime": 26.3565,
|
||
|
|
"eval_samples_per_second": 18.098,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 6800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02394904753365304,
|
||
|
|
"grad_norm": 0.8318554162979126,
|
||
|
|
"learning_rate": 2.394835485214494e-05,
|
||
|
|
"loss": 1.8293,
|
||
|
|
"step": 6900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02394904753365304,
|
||
|
|
"eval_loss": 1.80862295627594,
|
||
|
|
"eval_runtime": 26.4221,
|
||
|
|
"eval_samples_per_second": 18.053,
|
||
|
|
"eval_steps_per_second": 0.378,
|
||
|
|
"step": 6900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.024296135179068302,
|
||
|
|
"grad_norm": 0.9045419096946716,
|
||
|
|
"learning_rate": 2.4295432458697766e-05,
|
||
|
|
"loss": 1.8324,
|
||
|
|
"step": 7000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.024296135179068302,
|
||
|
|
"eval_loss": 1.805204153060913,
|
||
|
|
"eval_runtime": 26.4185,
|
||
|
|
"eval_samples_per_second": 18.056,
|
||
|
|
"eval_steps_per_second": 0.379,
|
||
|
|
"step": 7000
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 100,
|
||
|
|
"max_steps": 288111,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 200,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": false
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 1.46085192400896e+18,
|
||
|
|
"train_batch_size": 48,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|