Invalid JSON: Unexpected non-whitespace character after JSONat line 2, column 1
| {"eval_loss":1.0263168812,"eval_binary":0.0,"eval_rouge":0.8630774331,"eval_llm_as_a_judge":0.0,"eval_runtime":49.0875,"eval_samples_per_second":1.019,"eval_steps_per_second":1.019,"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":0.0247745516,"step":250,"loss":0.5324,"grad_norm":1.523032546,"learning_rate":0.0000060674,"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":0.0495491032,"step":500,"loss":0.1588,"grad_norm":1.861987114,"learning_rate":0.0000122585,"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":0.0743236547,"step":750,"loss":0.1432,"grad_norm":0.0460563935,"learning_rate":0.0000184497,"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":0.0990982063,"step":1000,"loss":0.1222,"grad_norm":0.7678105831,"learning_rate":0.0000246409,"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":0.1238727579,"step":1250,"loss":0.1289,"grad_norm":0.0857704207,"learning_rate":0.0000308321,"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":0.1486473095,"step":1500,"loss":0.123,"grad_norm":0.9776291847,"learning_rate":0.0000370233,"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":0.1734218611,"step":1750,"loss":0.1247,"grad_norm":0.6648195386,"learning_rate":0.0000432145,"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":0.1981964126,"step":2000,"loss":0.1222,"grad_norm":0.1074415967,"learning_rate":0.0000494056,"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":0.2229709642,"step":2250,"loss":0.1197,"grad_norm":0.4264952838,"learning_rate":0.0000497053,"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":0.2477455158,"step":2500,"loss":0.1319,"grad_norm":0.7932649851,"learning_rate":0.0000493793,"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":0.2725200674,"step":2750,"loss":0.124,"grad_norm":0.0169372503,"learning_rate":0.0000490533,"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":0.297294619,"step":3000,"loss":0.112,"grad_norm":1.0842401981,"learning_rate":0.0000487273,"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":0.3220691705,"step":3250,"loss":0.1105,"grad_norm":0.0589365289,"learning_rate":0.0000484014,"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":0.3468437221,"step":3500,"loss":0.117,"grad_norm":2.2370946407,"learning_rate":0.0000480754,"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":0.3716182737,"step":3750,"loss":0.1142,"grad_norm":0.2535243928,"learning_rate":0.0000477494,"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":0.3963928253,"step":4000,"loss":0.115,"grad_norm":0.7460672259,"learning_rate":0.0000474234,"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":0.4211673769,"step":4250,"loss":0.1315,"grad_norm":0.5937117338,"learning_rate":0.0000470974,"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":0.4459419285,"step":4500,"loss":0.1148,"grad_norm":0.0067167096,"learning_rate":0.0000467714,"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":0.47071648,"step":4750,"loss":0.1094,"grad_norm":0.6822088361,"learning_rate":0.0000464454,"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":0.4954910316,"step":5000,"loss":0.1123,"grad_norm":0.5732296705,"learning_rate":0.0000461194,"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":0.5202655832,"step":5250,"loss":0.105,"grad_norm":1.0334236622,"learning_rate":0.0000457935,"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":0.5450401348,"step":5500,"loss":0.1097,"grad_norm":0.1588985473,"learning_rate":0.0000454675,"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":0.5698146864,"step":5750,"loss":0.1125,"grad_norm":1.1788893938,"learning_rate":0.0000451415,"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":0.5945892379,"step":6000,"loss":0.1078,"grad_norm":0.230259642,"learning_rate":0.0000448155,"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":0.6193637895,"step":6250,"loss":0.1026,"grad_norm":0.9611893892,"learning_rate":0.0000444895,"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":0.6441383411,"step":6500,"loss":0.0974,"grad_norm":0.0160588399,"learning_rate":0.0000441635,"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":0.6689128927,"step":6750,"loss":0.0965,"grad_norm":0.3839409649,"learning_rate":0.0000438375,"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":0.6936874443,"step":7000,"loss":0.1098,"grad_norm":0.3105497658,"learning_rate":0.0000435115,"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":0.7184619958,"step":7250,"loss":0.1116,"grad_norm":0.0435747579,"learning_rate":0.0000431856,"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":0.7432365474,"step":7500,"loss":0.1061,"grad_norm":0.265334934,"learning_rate":0.0000428596,"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":0.768011099,"step":7750,"loss":0.1009,"grad_norm":0.647885263,"learning_rate":0.0000425336,"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":0.7927856506,"step":8000,"loss":0.0929,"grad_norm":1.2587280273,"learning_rate":0.0000422076,"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":0.8175602022,"step":8250,"loss":0.0992,"grad_norm":0.7607027888,"learning_rate":0.0000418816,"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":0.8423347537,"step":8500,"loss":0.097,"grad_norm":0.0234140866,"learning_rate":0.0000415556,"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":0.8671093053,"step":8750,"loss":0.0974,"grad_norm":1.6173541546,"learning_rate":0.0000412296,"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":0.8918838569,"step":9000,"loss":0.0992,"grad_norm":0.3484966755,"learning_rate":0.0000409036,"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":0.9166584085,"step":9250,"loss":0.1068,"grad_norm":0.4587571025,"learning_rate":0.0000405777,"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":0.9414329601,"step":9500,"loss":0.0989,"grad_norm":0.5602991581,"learning_rate":0.0000402517,"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":0.9662075116,"step":9750,"loss":0.0957,"grad_norm":0.3049614131,"learning_rate":0.0000399257,"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":0.9909820632,"step":10000,"loss":0.1069,"grad_norm":0.5346589684,"learning_rate":0.0000395997,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.0627685115,"eval_binary":0.5,"eval_rouge":0.9718670465,"eval_llm_as_a_judge":0.5,"eval_runtime":39.1908,"eval_samples_per_second":1.276,"eval_steps_per_second":1.276,"epoch":1.0,"step":10091,"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":1.0157566148,"step":10250,"loss":0.093,"grad_norm":1.1988984346,"learning_rate":0.0000392737,"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":1.0405311664,"step":10500,"loss":0.0856,"grad_norm":0.5537581444,"learning_rate":0.0000389477,"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":1.065305718,"step":10750,"loss":0.0932,"grad_norm":0.7537667155,"learning_rate":0.0000386217,"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":1.0900802695,"step":11000,"loss":0.0842,"grad_norm":0.5561412573,"learning_rate":0.0000382957,"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":1.1148548211,"step":11250,"loss":0.0867,"grad_norm":0.6610764861,"learning_rate":0.0000379711,"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":1.1396293727,"step":11500,"loss":0.08,"grad_norm":2.3849334717,"learning_rate":0.0000376451,"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":1.1644039243,"step":11750,"loss":0.0848,"grad_norm":0.4288468063,"learning_rate":0.0000373191,"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":1.1891784759,"step":12000,"loss":0.0825,"grad_norm":1.3528180122,"learning_rate":0.0000369944,"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":1.2139530275,"step":12250,"loss":0.0806,"grad_norm":1.9075659513,"learning_rate":0.0000366684,"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":1.238727579,"step":12500,"loss":0.0832,"grad_norm":0.8397055268,"learning_rate":0.0000363424,"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":1.2635021306,"step":12750,"loss":0.0877,"grad_norm":0.148780033,"learning_rate":0.0000360164,"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":1.2882766822,"step":13000,"loss":0.0866,"grad_norm":0.0096390331,"learning_rate":0.0000356904,"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":1.3130512338,"step":13250,"loss":0.0791,"grad_norm":0.0332148783,"learning_rate":0.0000353645,"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":1.3378257854,"step":13500,"loss":0.0809,"grad_norm":0.0060375775,"learning_rate":0.0000350385,"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":1.3626003369,"step":13750,"loss":0.0854,"grad_norm":0.0187670588,"learning_rate":0.0000347125,"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":1.3873748885,"step":14000,"loss":0.0922,"grad_norm":0.0784000084,"learning_rate":0.0000343865,"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":1.4121494401,"step":14250,"loss":0.0837,"grad_norm":0.0023212512,"learning_rate":0.0000340605,"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":1.4369239917,"step":14500,"loss":0.0848,"grad_norm":0.0057139895,"learning_rate":0.0000337345,"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":1.4616985433,"step":14750,"loss":0.0858,"grad_norm":0.5069279671,"learning_rate":0.0000334085,"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":1.4864730948,"step":15000,"loss":0.079,"grad_norm":0.7604624033,"learning_rate":0.0000330825,"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":1.5112476464,"step":15250,"loss":0.0929,"grad_norm":0.8304641247,"learning_rate":0.0000327566,"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":1.536022198,"step":15500,"loss":0.0883,"grad_norm":0.7251547575,"learning_rate":0.0000324319,"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":1.5607967496,"step":15750,"loss":0.0778,"grad_norm":1.3709897995,"learning_rate":0.0000321059,"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":1.5855713012,"step":16000,"loss":0.0802,"grad_norm":2.15670228,"learning_rate":0.0000317799,"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":1.6103458527,"step":16250,"loss":0.0831,"grad_norm":0.0983608514,"learning_rate":0.0000314539,"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":1.6351204043,"step":16500,"loss":0.0795,"grad_norm":0.4006144106,"learning_rate":0.0000311279,"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":1.6598949559,"step":16750,"loss":0.0856,"grad_norm":0.296091944,"learning_rate":0.0000308019,"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":1.6846695075,"step":17000,"loss":0.0889,"grad_norm":0.0206144322,"learning_rate":0.0000304759,"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":1.7094440591,"step":17250,"loss":0.0754,"grad_norm":0.7631092072,"learning_rate":0.00003015,"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":1.7342186106,"step":17500,"loss":0.0775,"grad_norm":0.670375824,"learning_rate":0.000029824,"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":1.7589931622,"step":17750,"loss":0.0817,"grad_norm":0.7563747168,"learning_rate":0.000029498,"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":1.7837677138,"step":18000,"loss":0.0876,"grad_norm":0.549282074,"learning_rate":0.000029172,"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":1.8085422654,"step":18250,"loss":0.0882,"grad_norm":0.3442634642,"learning_rate":0.000028846,"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":1.833316817,"step":18500,"loss":0.0861,"grad_norm":3.4848599434,"learning_rate":0.00002852,"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":1.8580913685,"step":18750,"loss":0.0921,"grad_norm":0.4657160044,"learning_rate":0.000028194,"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":1.8828659201,"step":19000,"loss":0.0792,"grad_norm":0.2213658243,"learning_rate":0.000027868,"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":1.9076404717,"step":19250,"loss":0.0849,"grad_norm":0.8199850321,"learning_rate":0.0000275421,"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":1.9324150233,"step":19500,"loss":0.0879,"grad_norm":0.0659607574,"learning_rate":0.0000272161,"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":1.9571895749,"step":19750,"loss":0.0855,"grad_norm":0.5881988406,"learning_rate":0.0000268914,"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":1.9819641264,"step":20000,"loss":0.0782,"grad_norm":0.5516498685,"learning_rate":0.0000265654,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.0534980074,"eval_binary":0.54,"eval_rouge":0.976516761,"eval_llm_as_a_judge":0.54,"eval_runtime":39.3735,"eval_samples_per_second":1.27,"eval_steps_per_second":1.27,"epoch":2.0,"step":20182,"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":2.006738678,"step":20250,"loss":0.0736,"grad_norm":0.7942800522,"learning_rate":0.0000262394,"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":2.0315132296,"step":20500,"loss":0.0626,"grad_norm":0.00088776,"learning_rate":0.0000259134,"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":2.0562877812,"step":20750,"loss":0.0667,"grad_norm":1.7306857109,"learning_rate":0.0000255874,"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":2.0810623328,"step":21000,"loss":0.0611,"grad_norm":0.0800337344,"learning_rate":0.0000252614,"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":2.1058368844,"step":21250,"loss":0.0651,"grad_norm":0.6958248019,"learning_rate":0.0000249355,"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":2.1306114359,"step":21500,"loss":0.0698,"grad_norm":0.6856549978,"learning_rate":0.0000246095,"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":2.1553859875,"step":21750,"loss":0.0619,"grad_norm":0.2869307995,"learning_rate":0.0000242835,"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":2.1801605391,"step":22000,"loss":0.0698,"grad_norm":0.2166688591,"learning_rate":0.0000239575,"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":2.2049350907,"step":22250,"loss":0.0668,"grad_norm":0.8755005598,"learning_rate":0.0000236315,"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":2.2297096423,"step":22500,"loss":0.0713,"grad_norm":0.2372335643,"learning_rate":0.0000233055,"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":2.2544841938,"step":22750,"loss":0.0593,"grad_norm":0.7852942944,"learning_rate":0.0000229795,"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":2.2792587454,"step":23000,"loss":0.067,"grad_norm":1.178085804,"learning_rate":0.0000226535,"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":2.304033297,"step":23250,"loss":0.0691,"grad_norm":0.7586830854,"learning_rate":0.0000223276,"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":2.3288078486,"step":23500,"loss":0.0673,"grad_norm":0.2725040615,"learning_rate":0.0000220016,"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":2.3535824002,"step":23750,"loss":0.0702,"grad_norm":0.8307171464,"learning_rate":0.0000216769,"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":2.3783569517,"step":24000,"loss":0.0693,"grad_norm":1.0476861,"learning_rate":0.0000213509,"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":2.4031315033,"step":24250,"loss":0.0601,"grad_norm":0.2545061409,"learning_rate":0.0000210249,"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":2.4279060549,"step":24500,"loss":0.0645,"grad_norm":0.5762472153,"learning_rate":0.0000206989,"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":2.4526806065,"step":24750,"loss":0.0754,"grad_norm":0.5169177055,"learning_rate":0.0000203729,"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":2.4774551581,"step":25000,"loss":0.0648,"grad_norm":0.3621724546,"learning_rate":0.0000200469,"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":2.5022297096,"step":25250,"loss":0.071,"grad_norm":0.9184062481,"learning_rate":0.000019721,"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":2.5270042612,"step":25500,"loss":0.0733,"grad_norm":0.056006413,"learning_rate":0.000019395,"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":2.5517788128,"step":25750,"loss":0.0667,"grad_norm":0.0178615637,"learning_rate":0.0000190703,"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":2.5765533644,"step":26000,"loss":0.0595,"grad_norm":1.1122207642,"learning_rate":0.0000187443,"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":2.601327916,"step":26250,"loss":0.0657,"grad_norm":0.0229800809,"learning_rate":0.0000184196,"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":2.6261024675,"step":26500,"loss":0.0718,"grad_norm":0.4926201105,"learning_rate":0.0000180936,"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":2.6508770191,"step":26750,"loss":0.0633,"grad_norm":0.6802681088,"learning_rate":0.0000177676,"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":2.6756515707,"step":27000,"loss":0.063,"grad_norm":0.3343955576,"learning_rate":0.0000174416,"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":2.7004261223,"step":27250,"loss":0.067,"grad_norm":0.0725091696,"learning_rate":0.0000171157,"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":2.7252006739,"step":27500,"loss":0.0678,"grad_norm":0.0133264381,"learning_rate":0.0000167897,"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":2.7499752254,"step":27750,"loss":0.0651,"grad_norm":0.4858928025,"learning_rate":0.0000164637,"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":2.774749777,"step":28000,"loss":0.0582,"grad_norm":0.0067026918,"learning_rate":0.0000161377,"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":2.7995243286,"step":28250,"loss":0.0682,"grad_norm":0.00316608,"learning_rate":0.0000158117,"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":2.8242988802,"step":28500,"loss":0.0648,"grad_norm":1.3722319603,"learning_rate":0.0000154857,"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":2.8490734318,"step":28750,"loss":0.0687,"grad_norm":0.9632564187,"learning_rate":0.0000151597,"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":2.8738479834,"step":29000,"loss":0.0567,"grad_norm":0.3386249244,"learning_rate":0.0000148337,"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":2.8986225349,"step":29250,"loss":0.0679,"grad_norm":0.3381203413,"learning_rate":0.0000145078,"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":2.9233970865,"step":29500,"loss":0.0609,"grad_norm":0.3560456634,"learning_rate":0.0000141818,"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":2.9481716381,"step":29750,"loss":0.0675,"grad_norm":0.0316419937,"learning_rate":0.0000138558,"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":2.9729461897,"step":30000,"loss":0.0637,"grad_norm":0.0011654273,"learning_rate":0.0000135298,"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":2.9977207413,"step":30250,"loss":0.0642,"grad_norm":0.0002235201,"learning_rate":0.0000132038,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.0562961921,"eval_binary":0.54,"eval_rouge":0.9778917608,"eval_llm_as_a_judge":0.54,"eval_runtime":39.6277,"eval_samples_per_second":1.262,"eval_steps_per_second":1.262,"epoch":3.0,"step":30273,"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":3.0224952928,"step":30500,"loss":0.0566,"grad_norm":0.7295501828,"learning_rate":0.0000128791,"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":3.0472698444,"step":30750,"loss":0.055,"grad_norm":2.9202334881,"learning_rate":0.0000125531,"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":3.072044396,"step":31000,"loss":0.0614,"grad_norm":0.2907790244,"learning_rate":0.0000122271,"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":3.0968189476,"step":31250,"loss":0.0532,"grad_norm":0.0033611483,"learning_rate":0.0000119012,"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":3.1215934992,"step":31500,"loss":0.0478,"grad_norm":0.5726149678,"learning_rate":0.0000115752,"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":3.1463680507,"step":31750,"loss":0.0544,"grad_norm":0.8363867402,"learning_rate":0.0000112492,"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":3.1711426023,"step":32000,"loss":0.0594,"grad_norm":0.7790529728,"learning_rate":0.0000109232,"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":3.1959171539,"step":32250,"loss":0.0544,"grad_norm":0.0100028552,"learning_rate":0.0000105972,"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":3.2206917055,"step":32500,"loss":0.0538,"grad_norm":0.0261486471,"learning_rate":0.0000102712,"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":3.2454662571,"step":32750,"loss":0.0492,"grad_norm":1.2334165573,"learning_rate":0.0000099452,"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":3.2702408086,"step":33000,"loss":0.0557,"grad_norm":0.4991805851,"learning_rate":0.0000096192,"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":3.2950153602,"step":33250,"loss":0.0522,"grad_norm":0.7711235285,"learning_rate":0.0000092946,"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":3.3197899118,"step":33500,"loss":0.047,"grad_norm":0.4563866556,"learning_rate":0.0000089686,"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":3.3445644634,"step":33750,"loss":0.0542,"grad_norm":0.9393787384,"learning_rate":0.0000086426,"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":3.369339015,"step":34000,"loss":0.0506,"grad_norm":0.209030211,"learning_rate":0.0000083166,"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":3.3941135665,"step":34250,"loss":0.0508,"grad_norm":0.5082834959,"learning_rate":0.0000079906,"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":3.4188881181,"step":34500,"loss":0.0474,"grad_norm":0.2087581158,"learning_rate":0.0000076646,"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":3.4436626697,"step":34750,"loss":0.0497,"grad_norm":0.2804997861,"learning_rate":0.0000073386,"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":3.4684372213,"step":35000,"loss":0.054,"grad_norm":0.8045966625,"learning_rate":0.0000070126,"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":3.4932117729,"step":35250,"loss":0.0463,"grad_norm":1.8549096584,"learning_rate":0.0000066867,"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":3.5179863244,"step":35500,"loss":0.0493,"grad_norm":0.0131222457,"learning_rate":0.0000063633,"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":3.542760876,"step":35750,"loss":0.049,"grad_norm":1.0247558355,"learning_rate":0.0000060373,"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":3.5675354276,"step":36000,"loss":0.0495,"grad_norm":1.1740553379,"learning_rate":0.0000057113,"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":3.5923099792,"step":36250,"loss":0.0537,"grad_norm":0.3666393459,"learning_rate":0.0000053853,"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":3.6170845308,"step":36500,"loss":0.0483,"grad_norm":1.2278428078,"learning_rate":0.0000050593,"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":3.6418590824,"step":36750,"loss":0.0507,"grad_norm":1.8344718218,"learning_rate":0.0000047333,"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":3.6666336339,"step":37000,"loss":0.0489,"grad_norm":1.3024396896,"learning_rate":0.0000044074,"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":3.6914081855,"step":37250,"loss":0.0529,"grad_norm":0.5916293859,"learning_rate":0.0000040814,"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":3.7161827371,"step":37500,"loss":0.0522,"grad_norm":0.796315372,"learning_rate":0.0000037554,"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":3.7409572887,"step":37750,"loss":0.0495,"grad_norm":1.1936062574,"learning_rate":0.0000034294,"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":3.7657318403,"step":38000,"loss":0.05,"grad_norm":0.0045746323,"learning_rate":0.0000031034,"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":3.7905063918,"step":38250,"loss":0.0525,"grad_norm":0.3080344498,"learning_rate":0.0000027774,"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":3.8152809434,"step":38500,"loss":0.0497,"grad_norm":0.7108872533,"learning_rate":0.0000024514,"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":3.840055495,"step":38750,"loss":0.0529,"grad_norm":0.5039321184,"learning_rate":0.0000021254,"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":3.8648300466,"step":39000,"loss":0.0515,"grad_norm":0.4520630538,"learning_rate":0.0000017995,"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":3.8896045982,"step":39250,"loss":0.0451,"grad_norm":0.5372409821,"learning_rate":0.0000014735,"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":3.9143791497,"step":39500,"loss":0.0579,"grad_norm":0.2379768789,"learning_rate":0.0000011475,"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":3.9391537013,"step":39750,"loss":0.0487,"grad_norm":0.878942132,"learning_rate":0.0000008228,"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":3.9639282529,"step":40000,"loss":0.0492,"grad_norm":0.1474274844,"learning_rate":0.0000004968,"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":3.9887028045,"step":40250,"loss":0.0538,"grad_norm":0.0074777692,"learning_rate":0.0000001708,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.0535237938,"eval_binary":0.54,"eval_rouge":0.9770745949,"eval_llm_as_a_judge":0.54,"eval_runtime":39.8893,"eval_samples_per_second":1.253,"eval_steps_per_second":1.253,"epoch":4.0,"step":40364,"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":40364,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":6734.6117,"train_samples_per_second":5.994,"train_steps_per_second":5.994,"total_flos":3.153073337e+17,"train_loss":0.0811412968} | |