eval-harness

reference

gold/ag_news_200_adj1 gold

Promoted as gold: the yardstick a decision rests on.
Promoted from person adjudication of the dataset's labels, 2026-10-19: promoted before labels existed (0.6.0): its tier was given when it was made

01what it is

tier
gold
dataset
ag_news_200
promoted from
person adjudication of the dataset's labels
why
promoted before labels existed (0.6.0): its tier was given when it was made
items
200
created
2026-10-19T15:20:38Z

02scored against it · 30

scored against silverscored by a judgewrote the reference it is scored againstscored against a label set, not a reference setunmarked rows are scored against gold

armmetricmeanitems
ag_anthropic_l_n200_v1 · adjudicatedcorrect0.985200inspect
ag_openai_m_n200_v1 · adjudicatedcorrect0.980200inspect
ag_openai_l_n200_v1 · adjudicatedcorrect0.975200inspect
ag_anthropic_m_n200_v1 · adjudicatedcorrect0.965200inspect
ag_llama_m_n200_v1 · adjudicatedcorrect0.965200inspect
ag_gemma_s_n200_v1 · adjudicatedcorrect0.960200inspect
ag_qwen_m_n200_v1 · adjudicatedcorrect0.960200inspect
ag_qwen_s_n200_v1 · adjudicatedcorrect0.960200inspect
ag_deepseek_l_n200_v1 · adjudicatedcorrect0.955200inspect
ag_gemma_l_n200_v1 · adjudicatedcorrect0.955200inspect
ag_qwen_l_n200_v1 · adjudicatedcorrect0.955200inspect
ag_mistral_l_n200_v1 · adjudicatedcorrect0.950200inspect
ag_glm_l_n200_v1 · adjudicatedcorrect0.945200inspect
ag_glm_m_n200_v1 · adjudicatedcorrect0.945200inspect
ag_mistral_m_n200_v1 · adjudicatedcorrect0.945200inspect
ag_openai_s_n200_v1 · adjudicatedcorrect0.945200inspect
ag_gemma_m_n200_v1 · adjudicatedcorrect0.940200inspect
ag_glm_s_n200_v1 · adjudicatedcorrect0.935200inspect
ag_mistral_s_n200_v1 · adjudicatedcorrect0.930200inspect
ag_anthropic_s_n200_v1 · adjudicatedcorrect0.925200inspect
ag_deepseek_m_n200_v1 · adjudicatedcorrect0.925200inspect
ag_deepseek_s_n200_v1 · adjudicatedcorrect0.925200inspect
ag_llama_l_n200_v1 · adjudicatedcorrect0.925200inspect
ag_llama_s_n200_v1 · adjudicatedcorrect0.920200inspect
ag_bert_base_fy_n200_v1 · adjudicatedcorrect0.905200inspect
ag_bert_base_ta_n200_v1 · adjudicatedcorrect0.905200inspect
ag_bert_mini_n200_v1 · adjudicatedcorrect0.905200inspect
ag_bart_mnli_n200_v1 · adjudicatedcorrect0.755200inspect
ag_keyword_n200_v1 · adjudicatedcorrect0.670200inspect
ag_constant_n200_v1 · adjudicatedcorrect0.225200inspect