eval-harness

model

eval-gpt-6sol llm

A model record: onboarded once for the project and used by any eval's arms.

weights
proprietary
aliases
eval-gpt-6sol
settings
defaults
onboarding
not onboarded: evh model-onboard
best result per goal
#1 of 33 A scorer that agrees with experts on summaries (SummEval)
#5 of 28 Named-entity extraction (Few-NERD)
#7 of 30 News topic classification (AG News)
#12 of 29 Short news summaries (CNN)
#16 of 28 Encyclopedia topic classification (DBpedia)

Every model

01deployments and their onboarding

No deployment is recorded for this model. Not onboarded yet: evh model-onboard eval-gpt-6sol.

02settings history

No change of settings or build is recorded.

03arms

armdeploymentgoalits settingsarm smokeresult
ag_openai_l_n200_v1 · adjudicated
ag_news_200
——max_tokens 200, temperature 0.0nonecorrect 0.975
cnn_openai_l_n200_v1 · judged/judge-summ-pick-v1
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonejudge_won 0.004
cnn_openai_l_n200_v1 · judged/judge-summ-v1
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonejudge_score 4.773
cnn_openai_l_n200_v1 · judged/judge-summ-v2
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonejudge_score 4.847
cnn_openai_l_n200_v1 · silver
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonecoverage 0.371
ag_openai_l_n200_v1
ag_news_200
—News topic classification (AG News)max_tokens 200, temperature 0.0nonecorrect 0.890
cnn_openai_l_n200_v1
cnn_dailymail_200
—Short news summaries (CNN)max_tokens 1200, temperature 0.0nonecoverage 0.327
db_openai_l_n200_v1
dbpedia_280
—Encyclopedia topic classification (DBpedia)max_tokens 200, temperature 0.0nonecorrect 0.975
fn_openai_l_n200_v1
few_nerd_280
—Named-entity extraction (Few-NERD)max_tokens 600, temperature 0.0nonef1 0.672
se_openai_l_n100_v1
summeval_100
—A scorer that agrees with experts on summaries (SummEval)max_tokens 1500, temperature 0.0nonetau 0.540