168 lines
62 KiB
JSON
168 lines
62 KiB
JSON
{"eval_loss":1.461114645,"eval_binary":0.0,"eval_rouge":0.840004776,"eval_llm_as_a_judge":0.0,"eval_runtime":38.5471,"eval_samples_per_second":1.297,"eval_steps_per_second":1.297,"epoch":0.0,"step":0,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.0247231013,"step":250,"loss":0.7402,"grad_norm":54.7997970581,"learning_rate":0.0000060059,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.0494462025,"step":500,"loss":0.3334,"grad_norm":2.3189218044,"learning_rate":0.0000121849,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.0741693038,"step":750,"loss":0.2563,"grad_norm":2.7480959892,"learning_rate":0.0000183638,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.0988924051,"step":1000,"loss":0.2416,"grad_norm":4.4334211349,"learning_rate":0.0000245428,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.1236155063,"step":1250,"loss":0.2444,"grad_norm":2.8457245827,"learning_rate":0.0000307217,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.1483386076,"step":1500,"loss":0.2232,"grad_norm":1.9762811661,"learning_rate":0.0000369006,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.1730617089,"step":1750,"loss":0.2032,"grad_norm":1.268830061,"learning_rate":0.0000430796,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.1977848101,"step":2000,"loss":0.21,"grad_norm":0.7073513865,"learning_rate":0.0000492585,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.2225079114,"step":2250,"loss":0.219,"grad_norm":0.1617899835,"learning_rate":0.0000497137,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.2472310127,"step":2500,"loss":0.1886,"grad_norm":1.803670764,"learning_rate":0.0000493884,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.2719541139,"step":2750,"loss":0.1907,"grad_norm":1.9092983007,"learning_rate":0.0000490631,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.2966772152,"step":3000,"loss":0.1858,"grad_norm":0.0111169936,"learning_rate":0.0000487378,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.3214003165,"step":3250,"loss":0.1831,"grad_norm":1.347692728,"learning_rate":0.0000484125,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.3461234177,"step":3500,"loss":0.1814,"grad_norm":1.0721465349,"learning_rate":0.0000480872,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.370846519,"step":3750,"loss":0.1719,"grad_norm":1.1927968264,"learning_rate":0.0000477619,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.3955696203,"step":4000,"loss":0.1701,"grad_norm":0.3336207867,"learning_rate":0.0000474366,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.4202927215,"step":4250,"loss":0.1657,"grad_norm":0.0028190741,"learning_rate":0.0000471113,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.4450158228,"step":4500,"loss":0.1542,"grad_norm":2.7712564468,"learning_rate":0.0000467859,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.4697389241,"step":4750,"loss":0.1612,"grad_norm":1.4837995768,"learning_rate":0.0000464606,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.4944620253,"step":5000,"loss":0.1543,"grad_norm":0.0042050271,"learning_rate":0.0000461353,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.5191851266,"step":5250,"loss":0.1559,"grad_norm":1.55856359,"learning_rate":0.00004581,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.5439082278,"step":5500,"loss":0.1626,"grad_norm":0.7253758311,"learning_rate":0.0000454847,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.5686313291,"step":5750,"loss":0.1566,"grad_norm":0.3586200476,"learning_rate":0.0000451594,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.5933544304,"step":6000,"loss":0.1529,"grad_norm":0.2762477696,"learning_rate":0.0000448341,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.6180775316,"step":6250,"loss":0.1441,"grad_norm":0.0548379533,"learning_rate":0.0000445088,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.6428006329,"step":6500,"loss":0.1533,"grad_norm":1.5881824493,"learning_rate":0.0000441835,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.6675237342,"step":6750,"loss":0.1371,"grad_norm":0.6844435334,"learning_rate":0.0000438582,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.6922468354,"step":7000,"loss":0.1543,"grad_norm":1.086692214,"learning_rate":0.0000435329,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.7169699367,"step":7250,"loss":0.1434,"grad_norm":1.3913636208,"learning_rate":0.0000432075,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.741693038,"step":7500,"loss":0.1319,"grad_norm":1.0691365004,"learning_rate":0.0000428822,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.7664161392,"step":7750,"loss":0.1416,"grad_norm":8.5588874817,"learning_rate":0.0000425569,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.7911392405,"step":8000,"loss":0.1373,"grad_norm":0.0045287628,"learning_rate":0.0000422316,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.8158623418,"step":8250,"loss":0.1662,"grad_norm":0.4163950086,"learning_rate":0.0000419063,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.840585443,"step":8500,"loss":0.1272,"grad_norm":2.7291164398,"learning_rate":0.000041581,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.8653085443,"step":8750,"loss":0.1431,"grad_norm":1.3981182575,"learning_rate":0.0000412557,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.8900316456,"step":9000,"loss":0.1534,"grad_norm":0.3506095409,"learning_rate":0.0000409304,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.9147547468,"step":9250,"loss":0.1387,"grad_norm":1.2084834576,"learning_rate":0.0000406051,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.9394778481,"step":9500,"loss":0.1349,"grad_norm":1.2688090801,"learning_rate":0.0000402798,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.9642009494,"step":9750,"loss":0.1288,"grad_norm":0.8546839952,"learning_rate":0.0000399545,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.9889240506,"step":10000,"loss":0.1232,"grad_norm":0.5538005829,"learning_rate":0.0000396291,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":0.1339163333,"eval_binary":0.4,"eval_rouge":0.9484506499,"eval_llm_as_a_judge":0.4,"eval_runtime":37.5908,"eval_samples_per_second":1.33,"eval_steps_per_second":1.33,"epoch":1.0,"step":10112,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.0136471519,"step":10250,"loss":0.1237,"grad_norm":0.9612093568,"learning_rate":0.0000393038,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.0383702532,"step":10500,"loss":0.1067,"grad_norm":1.7282613516,"learning_rate":0.0000389785,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.0630933544,"step":10750,"loss":0.1115,"grad_norm":0.8479132056,"learning_rate":0.0000386532,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.0878164557,"step":11000,"loss":0.1068,"grad_norm":0.9790354967,"learning_rate":0.0000383279,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.112539557,"step":11250,"loss":0.1144,"grad_norm":1.2608872652,"learning_rate":0.0000380026,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.1372626582,"step":11500,"loss":0.1143,"grad_norm":1.1066915989,"learning_rate":0.0000376773,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.1619857595,"step":11750,"loss":0.1049,"grad_norm":1.5078562498,"learning_rate":0.000037352,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.1867088608,"step":12000,"loss":0.1134,"grad_norm":0.020030342,"learning_rate":0.0000370267,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.211431962,"step":12250,"loss":0.1165,"grad_norm":1.020694375,"learning_rate":0.0000367014,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.2361550633,"step":12500,"loss":0.1143,"grad_norm":0.6087476611,"learning_rate":0.0000363774,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.2608781646,"step":12750,"loss":0.11,"grad_norm":0.2116389126,"learning_rate":0.000036052,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.2856012658,"step":13000,"loss":0.1087,"grad_norm":1.488263011,"learning_rate":0.0000357267,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.3103243671,"step":13250,"loss":0.1105,"grad_norm":1.8852481842,"learning_rate":0.0000354014,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.3350474684,"step":13500,"loss":0.1006,"grad_norm":0.6119297147,"learning_rate":0.0000350761,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.3597705696,"step":13750,"loss":0.114,"grad_norm":0.0027765117,"learning_rate":0.0000347508,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.3844936709,"step":14000,"loss":0.119,"grad_norm":1.5043395758,"learning_rate":0.0000344255,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.4092167722,"step":14250,"loss":0.1142,"grad_norm":0.0535508208,"learning_rate":0.0000341002,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.4339398734,"step":14500,"loss":0.1183,"grad_norm":0.336640507,"learning_rate":0.0000337749,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.4586629747,"step":14750,"loss":0.1165,"grad_norm":0.9554659724,"learning_rate":0.0000334496,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.4833860759,"step":15000,"loss":0.1115,"grad_norm":0.4799798429,"learning_rate":0.0000331243,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.5081091772,"step":15250,"loss":0.1158,"grad_norm":0.4781483412,"learning_rate":0.000032799,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.5328322785,"step":15500,"loss":0.1084,"grad_norm":0.4871940613,"learning_rate":0.0000324736,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.5575553797,"step":15750,"loss":0.1099,"grad_norm":1.3032745123,"learning_rate":0.0000321483,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.582278481,"step":16000,"loss":0.1211,"grad_norm":0.6182886362,"learning_rate":0.000031823,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.6070015823,"step":16250,"loss":0.1175,"grad_norm":0.526489675,"learning_rate":0.0000314977,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.6317246835,"step":16500,"loss":0.1062,"grad_norm":1.750696063,"learning_rate":0.0000311724,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.6564477848,"step":16750,"loss":0.1004,"grad_norm":1.6210639477,"learning_rate":0.0000308471,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.6811708861,"step":17000,"loss":0.0958,"grad_norm":0.5644216537,"learning_rate":0.0000305218,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.7058939873,"step":17250,"loss":0.1147,"grad_norm":1.8764464855,"learning_rate":0.0000301965,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.7306170886,"step":17500,"loss":0.1126,"grad_norm":3.0712196827,"learning_rate":0.0000298712,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.7553401899,"step":17750,"loss":0.1019,"grad_norm":0.000978192,"learning_rate":0.0000295459,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.7800632911,"step":18000,"loss":0.1243,"grad_norm":2.5946235657,"learning_rate":0.0000292219,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.8047863924,"step":18250,"loss":0.1077,"grad_norm":2.3117940426,"learning_rate":0.0000288966,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.8295094937,"step":18500,"loss":0.1103,"grad_norm":0.3653891087,"learning_rate":0.0000285725,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.8542325949,"step":18750,"loss":0.1004,"grad_norm":1.3803018332,"learning_rate":0.0000282472,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.8789556962,"step":19000,"loss":0.1112,"grad_norm":2.6601779461,"learning_rate":0.0000279219,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.9036787975,"step":19250,"loss":0.109,"grad_norm":0.009044609,"learning_rate":0.0000275966,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.9284018987,"step":19500,"loss":0.1026,"grad_norm":0.2560603917,"learning_rate":0.0000272713,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.953125,"step":19750,"loss":0.1221,"grad_norm":1.5156265497,"learning_rate":0.000026946,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.9778481013,"step":20000,"loss":0.1063,"grad_norm":0.0502424724,"learning_rate":0.0000266207,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":0.1202029735,"eval_binary":0.44,"eval_rouge":0.9528717907,"eval_llm_as_a_judge":0.44,"eval_runtime":37.6019,"eval_samples_per_second":1.33,"eval_steps_per_second":1.33,"epoch":2.0,"step":20224,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.0025712025,"step":20250,"loss":0.0969,"grad_norm":1.1122921705,"learning_rate":0.0000262954,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.0272943038,"step":20500,"loss":0.0884,"grad_norm":0.928889513,"learning_rate":0.0000259701,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.0520174051,"step":20750,"loss":0.0899,"grad_norm":0.8800064921,"learning_rate":0.0000256448,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.0767405063,"step":21000,"loss":0.0919,"grad_norm":1.1201504469,"learning_rate":0.0000253195,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.1014636076,"step":21250,"loss":0.0962,"grad_norm":1.5560466051,"learning_rate":0.0000249941,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.1261867089,"step":21500,"loss":0.0907,"grad_norm":0.5000078082,"learning_rate":0.0000246688,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.1509098101,"step":21750,"loss":0.0879,"grad_norm":0.5242018104,"learning_rate":0.0000243435,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.1756329114,"step":22000,"loss":0.0866,"grad_norm":1.4044090509,"learning_rate":0.0000240182,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2003560127,"step":22250,"loss":0.0964,"grad_norm":0.8841950893,"learning_rate":0.0000236929,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2250791139,"step":22500,"loss":0.0854,"grad_norm":0.8067905903,"learning_rate":0.0000233676,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2498022152,"step":22750,"loss":0.0828,"grad_norm":1.169241786,"learning_rate":0.0000230423,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2745253165,"step":23000,"loss":0.0928,"grad_norm":0.1877602637,"learning_rate":0.0000227183,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2992484177,"step":23250,"loss":0.0948,"grad_norm":0.0010576567,"learning_rate":0.000022393,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.323971519,"step":23500,"loss":0.0839,"grad_norm":0.0008563405,"learning_rate":0.0000220677,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.3486946203,"step":23750,"loss":0.0949,"grad_norm":1.6228750944,"learning_rate":0.0000217424,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.3734177215,"step":24000,"loss":0.0896,"grad_norm":1.3844517469,"learning_rate":0.000021417,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.3981408228,"step":24250,"loss":0.0858,"grad_norm":1.2112687826,"learning_rate":0.0000210917,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.4228639241,"step":24500,"loss":0.0977,"grad_norm":1.0013253689,"learning_rate":0.0000207664,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.4475870253,"step":24750,"loss":0.0766,"grad_norm":0.000829512,"learning_rate":0.0000204411,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.4723101266,"step":25000,"loss":0.0784,"grad_norm":1.2415456772,"learning_rate":0.0000201158,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.4970332278,"step":25250,"loss":0.0862,"grad_norm":0.113463819,"learning_rate":0.0000197905,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.5217563291,"step":25500,"loss":0.082,"grad_norm":2.2702343464,"learning_rate":0.0000194652,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.5464794304,"step":25750,"loss":0.0827,"grad_norm":0.0194308143,"learning_rate":0.0000191399,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.5712025316,"step":26000,"loss":0.0857,"grad_norm":0.4198535383,"learning_rate":0.0000188146,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.5959256329,"step":26250,"loss":0.0919,"grad_norm":2.4839465618,"learning_rate":0.0000184893,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.6206487342,"step":26500,"loss":0.085,"grad_norm":0.4629532397,"learning_rate":0.000018164,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.6453718354,"step":26750,"loss":0.0828,"grad_norm":0.2432839572,"learning_rate":0.0000178386,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.6700949367,"step":27000,"loss":0.0881,"grad_norm":1.1563462019,"learning_rate":0.0000175146,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.694818038,"step":27250,"loss":0.0845,"grad_norm":1.5228748322,"learning_rate":0.0000171893,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.7195411392,"step":27500,"loss":0.0828,"grad_norm":0.5037127137,"learning_rate":0.000016864,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.7442642405,"step":27750,"loss":0.0914,"grad_norm":2.4526147842,"learning_rate":0.0000165387,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.7689873418,"step":28000,"loss":0.0946,"grad_norm":0.6520683765,"learning_rate":0.0000162147,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.793710443,"step":28250,"loss":0.081,"grad_norm":0.5551896691,"learning_rate":0.0000158894,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.8184335443,"step":28500,"loss":0.089,"grad_norm":0.3599557281,"learning_rate":0.0000155641,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.8431566456,"step":28750,"loss":0.0762,"grad_norm":1.3633737564,"learning_rate":0.0000152388,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.8678797468,"step":29000,"loss":0.0848,"grad_norm":0.0641759932,"learning_rate":0.0000149135,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.8926028481,"step":29250,"loss":0.089,"grad_norm":2.0974826813,"learning_rate":0.0000145882,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.9173259494,"step":29500,"loss":0.0736,"grad_norm":1.4356244802,"learning_rate":0.0000142628,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.9420490506,"step":29750,"loss":0.083,"grad_norm":1.1658858061,"learning_rate":0.0000139375,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.9667721519,"step":30000,"loss":0.0893,"grad_norm":0.3766824603,"learning_rate":0.0000136135,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.9914952532,"step":30250,"loss":0.089,"grad_norm":0.3294841349,"learning_rate":0.0000132882,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":0.0955977216,"eval_binary":0.48,"eval_rouge":0.9629392864,"eval_llm_as_a_judge":0.48,"eval_runtime":34.8289,"eval_samples_per_second":1.436,"eval_steps_per_second":1.436,"epoch":3.0,"step":30336,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.0162183544,"step":30500,"loss":0.0705,"grad_norm":1.3704227209,"learning_rate":0.0000129629,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.0409414557,"step":30750,"loss":0.0662,"grad_norm":0.5225652456,"learning_rate":0.0000126376,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.065664557,"step":31000,"loss":0.0739,"grad_norm":0.0499449074,"learning_rate":0.0000123123,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.0903876582,"step":31250,"loss":0.0819,"grad_norm":0.2583652139,"learning_rate":0.000011987,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.1151107595,"step":31500,"loss":0.0867,"grad_norm":0.0035025831,"learning_rate":0.0000116617,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.1398338608,"step":31750,"loss":0.0743,"grad_norm":0.6238996983,"learning_rate":0.0000113364,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.164556962,"step":32000,"loss":0.0794,"grad_norm":0.7921465039,"learning_rate":0.0000110124,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.1892800633,"step":32250,"loss":0.0757,"grad_norm":0.0061428845,"learning_rate":0.0000106884,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.2140031646,"step":32500,"loss":0.073,"grad_norm":1.9250328541,"learning_rate":0.000010363,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.2387262658,"step":32750,"loss":0.0686,"grad_norm":1.1742448807,"learning_rate":0.0000100377,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.2634493671,"step":33000,"loss":0.0724,"grad_norm":0.0734039173,"learning_rate":0.0000097124,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.2881724684,"step":33250,"loss":0.0625,"grad_norm":1.8306051493,"learning_rate":0.0000093871,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.3128955696,"step":33500,"loss":0.065,"grad_norm":0.6452906132,"learning_rate":0.0000090618,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.3376186709,"step":33750,"loss":0.0672,"grad_norm":1.2680094242,"learning_rate":0.0000087365,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.3623417722,"step":34000,"loss":0.071,"grad_norm":0.0446249433,"learning_rate":0.0000084112,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.3870648734,"step":34250,"loss":0.0694,"grad_norm":2.3345537186,"learning_rate":0.0000080859,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4117879747,"step":34500,"loss":0.0741,"grad_norm":0.3106979132,"learning_rate":0.0000077606,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4365110759,"step":34750,"loss":0.0778,"grad_norm":3.5347332954,"learning_rate":0.0000074353,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4612341772,"step":35000,"loss":0.0611,"grad_norm":2.0303833485,"learning_rate":0.00000711,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4859572785,"step":35250,"loss":0.0773,"grad_norm":0.0415618494,"learning_rate":0.0000067846,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.5106803797,"step":35500,"loss":0.0828,"grad_norm":1.2193745375,"learning_rate":0.0000064593,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.535403481,"step":35750,"loss":0.0792,"grad_norm":1.887878418,"learning_rate":0.000006134,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.5601265823,"step":36000,"loss":0.0681,"grad_norm":1.181163311,"learning_rate":0.0000058087,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.5848496835,"step":36250,"loss":0.0717,"grad_norm":1.101999402,"learning_rate":0.0000054834,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.6095727848,"step":36500,"loss":0.0772,"grad_norm":1.6269752979,"learning_rate":0.0000051581,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.6342958861,"step":36750,"loss":0.0666,"grad_norm":1.4130914211,"learning_rate":0.0000048328,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.6590189873,"step":37000,"loss":0.0719,"grad_norm":1.5563285351,"learning_rate":0.0000045088,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.6837420886,"step":37250,"loss":0.0637,"grad_norm":0.5765205622,"learning_rate":0.0000041835,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.7084651899,"step":37500,"loss":0.0693,"grad_norm":0.3765734434,"learning_rate":0.0000038595,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.7331882911,"step":37750,"loss":0.0637,"grad_norm":0.5553440452,"learning_rate":0.0000035342,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.7579113924,"step":38000,"loss":0.0627,"grad_norm":1.6216650009,"learning_rate":0.0000032088,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.7826344937,"step":38250,"loss":0.0656,"grad_norm":4.0614652634,"learning_rate":0.0000028835,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.8073575949,"step":38500,"loss":0.0646,"grad_norm":0.9978199601,"learning_rate":0.0000025582,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.8320806962,"step":38750,"loss":0.065,"grad_norm":1.2803815603,"learning_rate":0.0000022329,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.8568037975,"step":39000,"loss":0.0609,"grad_norm":1.7353192568,"learning_rate":0.0000019076,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.8815268987,"step":39250,"loss":0.073,"grad_norm":1.3712269068,"learning_rate":0.0000015823,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.90625,"step":39500,"loss":0.0771,"grad_norm":0.9225087762,"learning_rate":0.000001257,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.9309731013,"step":39750,"loss":0.0698,"grad_norm":0.0060281986,"learning_rate":0.0000009317,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.9556962025,"step":40000,"loss":0.0763,"grad_norm":1.3945635557,"learning_rate":0.0000006064,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.9804193038,"step":40250,"loss":0.0648,"grad_norm":2.6979146004,"learning_rate":0.0000002811,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":0.1107318476,"eval_binary":0.46,"eval_rouge":0.9610821866,"eval_llm_as_a_judge":0.46,"eval_runtime":36.7743,"eval_samples_per_second":1.36,"eval_steps_per_second":1.36,"epoch":4.0,"step":40448,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null}
|
|
{"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":4.0,"step":40448,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":4982.372,"train_samples_per_second":8.118,"train_steps_per_second":8.118,"total_flos":4.056404094e+16,"train_loss":0.1136261352}
|