eval-harness

reference

gold/ag_news_200 gold

Promoted as gold: the yardstick a decision rests on.
Promoted from person the dataset's authors, 2026-09-27: 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 the dataset's authors
why
promoted before labels existed (0.6.0): its tier was given when it was made
items
200
created
2026-09-27T08:02:32Z

02scored against it · 33

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_bert_base_ta_n200_v1correct0.960200inspect
ag_bert_base_ta_n200_v1 · linuxcorrect0.960200inspect
ag_bert_base_fy_n200_v1correct0.950200inspect
ag_bert_base_fy_n200_v1 · linuxcorrect0.950200inspect
ag_bert_mini_n200_v1correct0.945200inspect
ag_bert_mini_n200_v1 · linuxcorrect0.945200inspect
ag_anthropic_m_n200_v1correct0.910200inspect
ag_anthropic_l_n200_v1correct0.900200inspect
ag_openai_m_n200_v1correct0.895200inspect
ag_openai_l_n200_v1correct0.890200inspect
ag_qwen_m_n200_v1correct0.885200inspect
ag_llama_m_n200_v1correct0.880200inspect
ag_gemma_s_n200_v1correct0.875200inspect
ag_mistral_l_n200_v1correct0.875200inspect
ag_qwen_s_n200_v1correct0.875200inspect
ag_deepseek_l_n200_v1correct0.870200inspect
ag_gemma_l_n200_v1correct0.870200inspect
ag_qwen_l_n200_v1correct0.870200inspect
ag_glm_l_n200_v1correct0.860200inspect
ag_glm_m_n200_v1correct0.860200inspect
ag_mistral_m_n200_v1correct0.860200inspect
ag_mistral_s_n200_v1correct0.860200inspect
ag_openai_s_n200_v1correct0.860200inspect
ag_gemma_m_n200_v1correct0.855200inspect
ag_glm_s_n200_v1correct0.850200inspect
ag_deepseek_m_n200_v1correct0.845200inspect
ag_anthropic_s_n200_v1correct0.840200inspect
ag_deepseek_s_n200_v1correct0.840200inspect
ag_llama_l_n200_v1correct0.840200inspect
ag_llama_s_n200_v1correct0.835200inspect
ag_bart_mnli_n200_v1correct0.700200inspect
ag_keyword_n200_v1correct0.670200inspect
ag_constant_n200_v1correct0.250200inspect