eval-harness

model

eval-glm-5 llm

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

weights
open
licence
MIT
aliases
eval-glm-5
settings
defaults
onboarding
not onboarded: evh model-onboard
best result per goal
#9 of 33 A scorer that agrees with experts on summaries (SummEval)
#16 of 30 News topic classification (AG News)
#18 of 28 Encyclopedia topic classification (DBpedia)
#21 of 29 Short news summaries (CNN)
#22 of 28 Named-entity extraction (Few-NERD)

Every model

01deployments and their onboarding

No deployment is recorded for this model. Not onboarded yet: evh model-onboard eval-glm-5.

02settings history

No change of settings or build is recorded.

03arms

armdeploymentgoalits settingsarm smokeresult
ag_glm_l_n200_v1 · adjudicated
ag_news_200
——max_tokens 200, temperature 0.0nonecorrect 0.945
cnn_glm_l_n200_v1 · judged/judge-summ-pick-v1
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonejudge_won 0.006
cnn_glm_l_n200_v1 · judged/judge-summ-v1
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonejudge_score 4.607
cnn_glm_l_n200_v1 · judged/judge-summ-v2
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonejudge_score 4.384
cnn_glm_l_n200_v1 · silver
cnn_dailymail_200
——max_tokens 1200, temperature 0.0nonecoverage 0.375
ag_glm_l_n200_v1
ag_news_200
—News topic classification (AG News)max_tokens 200, temperature 0.0nonecorrect 0.860
cnn_glm_l_n200_v1
cnn_dailymail_200
—Short news summaries (CNN)max_tokens 1200, temperature 0.0nonecoverage 0.312
db_glm_l_n200_v1
dbpedia_280
—Encyclopedia topic classification (DBpedia)max_tokens 200, temperature 0.0nonecorrect 0.968
fn_glm_l_n200_v1
few_nerd_280
—Named-entity extraction (Few-NERD)max_tokens 600, temperature 0.0nonef1 0.566
se_glm_l_n100_v1
summeval_100
—A scorer that agrees with experts on summaries (SummEval)max_tokens 1500, temperature 0.0nonetau 0.492