eval-harness

metric

judge_gemma_l

better
higher
kind
quality: measured against a reference or by a judge
decides
no goal
weighed by
no goal
recorded on
cnn_dailymail_200

01what it measures, and how

Judge gemma_l's mean score for the output, on the scorer's scale.

A project explains its own metric at length with METRIC_DESCRIPTIONS = {"judge_gemma_l": {"summary": "…", "details": "…"}}.

02on cnn_dailymail_200

judge judge-summ-v1cnn_anthropic_l_n200_v1 · judged/judge-summ-v1: 4.995cnn_anthropic_m_n200_v1 · judged/judge-summ-v1: 4.985cnn_anthropic_s_n200_v1 · judged/judge-summ-v1: 4.976cnn_deepseek_m_n200_v1 · judged/judge-summ-v1: 4.967cnn_deepseek_l_n200_v1 · judged/judge-summ-v1: 4.937cnn_glm_m_n200_v1 · judged/judge-summ-v1: 4.937cnn_qwen_m_n200_v1 · judged/judge-summ-v1: 4.923cnn_deepseek_s_n200_v1 · judged/judge-summ-v1: 4.906cnn_mistral_s_n200_v1 · judged/judge-summ-v1: 4.902cnn_gemma_s_n200_v1 · judged/judge-summ-v1: 4.901cnn_openai_s_n200_v1 · judged/judge-summ-v1: 4.897cnn_mistral_l_n200_v1 · judged/judge-summ-v1: 4.887cnn_openai_m_n200_v1 · judged/judge-summ-v1: 4.877cnn_openai_l_n200_v1 · judged/judge-summ-v1: 4.869cnn_gemma_l_n200_v1 · judged/judge-summ-v1: 4.855cnn_qwen_l_n200_v1 · judged/judge-summ-v1: 4.85cnn_llama_l_n200_v1 · judged/judge-summ-v1: 4.84cnn_mistral_m_n200_v1 · judged/judge-summ-v1: 4.84cnn_llama_s_n200_v1 · judged/judge-summ-v1: 4.832cnn_llama_m_n200_v1 · judged/judge-summ-v1: 4.829cnn_gemma_m_n200_v1 · judged/judge-summ-v1: 4.827cnn_qwen_s_n200_v1 · judged/judge-summ-v1: 4.73cnn_glm_s_n200_v1 · judged/judge-summ-v1: 4.722cnn_glm_l_n200_v1 · judged/judge-summ-v1: 4.678cnn_bart_l_n200_v1 · judged/judge-summ-v1: 4.129cnn_bart_m_n200_v1 · judged/judge-summ-v1: 3.957cnn_bart_l_xsum_n200_v1 · judged/judge-summ-v1: 3.887cnn_lead3_n200_v1 · judged/judge-summ-v1: 3.721cnn_bart_s_n200_v1 · judged/judge-summ-v1: 3.542judge judge-summ-pick-v1cnn_anthropic_l_n200_v1 · judged/judge-summ-pick-v1: 0.385cnn_anthropic_s_n200_v1 · judged/judge-summ-pick-v1: 0.241cnn_anthropic_m_n200_v1 · judged/judge-summ-pick-v1: 0.146cnn_deepseek_m_n200_v1 · judged/judge-summ-pick-v1: 0.091cnn_qwen_l_n200_v1 · judged/judge-summ-pick-v1: 0.048cnn_qwen_m_n200_v1 · judged/judge-summ-pick-v1: 0.015cnn_glm_l_n200_v1 · judged/judge-summ-pick-v1: 0.013cnn_glm_m_n200_v1 · judged/judge-summ-pick-v1: 0.011cnn_mistral_s_n200_v1 · judged/judge-summ-pick-v1: 0.0076cnn_deepseek_l_n200_v1 · judged/judge-summ-pick-v1: 0.0073cnn_deepseek_s_n200_v1 · judged/judge-summ-pick-v1: 0.0073cnn_gemma_m_n200_v1 · judged/judge-summ-pick-v1: 0.0058cnn_openai_s_n200_v1 · judged/judge-summ-pick-v1: 0.0055cnn_openai_l_n200_v1 · judged/judge-summ-pick-v1: 0.0038cnn_gemma_s_n200_v1 · judged/judge-summ-pick-v1: 0.0031cnn_llama_s_n200_v1 · judged/judge-summ-pick-v1: 0.0025cnn_qwen_s_n200_v1 · judged/judge-summ-pick-v1: 0.0022cnn_glm_s_n200_v1 · judged/judge-summ-pick-v1: 0.002cnn_openai_m_n200_v1 · judged/judge-summ-pick-v1: 0.002cnn_gemma_l_n200_v1 · judged/judge-summ-pick-v1: 0.0013cnn_llama_m_n200_v1 · judged/judge-summ-pick-v1: 0.0013cnn_llama_l_n200_v1 · judged/judge-summ-pick-v1: 0.00084cnn_bart_l_n200_v1 · judged/judge-summ-pick-v1: 0cnn_bart_l_xsum_n200_v1 · judged/judge-summ-pick-v1: 0cnn_bart_m_n200_v1 · judged/judge-summ-pick-v1: 0cnn_bart_s_n200_v1 · judged/judge-summ-pick-v1: 0cnn_lead3_n200_v1 · judged/judge-summ-pick-v1: 0cnn_mistral_l_n200_v1 · judged/judge-summ-pick-v1: 0cnn_mistral_m_n200_v1 · judged/judge-summ-pick-v1: 00better · 4.995

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

armmean judge_gemma_litemsscored against
cnn_anthropic_l_n200_v1 · judged/judge-summ-v14.995197judge judge-summ-v1
cnn_anthropic_m_n200_v1 · judged/judge-summ-v14.985197judge judge-summ-v1
cnn_anthropic_s_n200_v1 · judged/judge-summ-v14.976197judge judge-summ-v1
cnn_deepseek_m_n200_v1 · judged/judge-summ-v14.967197judge judge-summ-v1
cnn_deepseek_l_n200_v1 · judged/judge-summ-v14.937197judge judge-summ-v1
cnn_glm_m_n200_v1 · judged/judge-summ-v14.937197judge judge-summ-v1
cnn_qwen_m_n200_v1 · judged/judge-summ-v14.923197judge judge-summ-v1
cnn_deepseek_s_n200_v1 · judged/judge-summ-v14.906197judge judge-summ-v1
cnn_mistral_s_n200_v1 · judged/judge-summ-v14.902197judge judge-summ-v1
cnn_gemma_s_n200_v1 · judged/judge-summ-v14.901197judge judge-summ-v1
cnn_openai_s_n200_v1 · judged/judge-summ-v14.897197judge judge-summ-v1
cnn_mistral_l_n200_v1 · judged/judge-summ-v14.887197judge judge-summ-v1
cnn_openai_m_n200_v1 · judged/judge-summ-v14.877197judge judge-summ-v1
cnn_openai_l_n200_v1 · judged/judge-summ-v14.869197judge judge-summ-v1
cnn_gemma_l_n200_v1 · judged/judge-summ-v14.855197judge judge-summ-v1
cnn_qwen_l_n200_v1 · judged/judge-summ-v14.850197judge judge-summ-v1
cnn_llama_l_n200_v1 · judged/judge-summ-v14.840197judge judge-summ-v1
cnn_mistral_m_n200_v1 · judged/judge-summ-v14.840197judge judge-summ-v1
cnn_llama_s_n200_v1 · judged/judge-summ-v14.832197judge judge-summ-v1
cnn_llama_m_n200_v1 · judged/judge-summ-v14.829197judge judge-summ-v1
cnn_gemma_m_n200_v1 · judged/judge-summ-v14.827197judge judge-summ-v1
cnn_qwen_s_n200_v1 · judged/judge-summ-v14.730197judge judge-summ-v1
cnn_glm_s_n200_v1 · judged/judge-summ-v14.722197judge judge-summ-v1
cnn_glm_l_n200_v1 · judged/judge-summ-v14.678197judge judge-summ-v1
cnn_bart_l_n200_v1 · judged/judge-summ-v14.129197judge judge-summ-v1
cnn_bart_m_n200_v1 · judged/judge-summ-v13.957197judge judge-summ-v1
cnn_bart_l_xsum_n200_v1 · judged/judge-summ-v13.887197judge judge-summ-v1
cnn_lead3_n200_v1 · judged/judge-summ-v13.721197judge judge-summ-v1
cnn_bart_s_n200_v1 · judged/judge-summ-v13.542197judge judge-summ-v1
cnn_anthropic_l_n200_v1 · judged/judge-summ-pick-v10.385198judge judge-summ-pick-v1
cnn_anthropic_s_n200_v1 · judged/judge-summ-pick-v10.241198judge judge-summ-pick-v1
cnn_anthropic_m_n200_v1 · judged/judge-summ-pick-v10.146198judge judge-summ-pick-v1
cnn_deepseek_m_n200_v1 · judged/judge-summ-pick-v10.091198judge judge-summ-pick-v1
cnn_qwen_l_n200_v1 · judged/judge-summ-pick-v10.048198judge judge-summ-pick-v1
cnn_qwen_m_n200_v1 · judged/judge-summ-pick-v10.015198judge judge-summ-pick-v1
cnn_glm_l_n200_v1 · judged/judge-summ-pick-v10.013198judge judge-summ-pick-v1
cnn_glm_m_n200_v1 · judged/judge-summ-pick-v10.011198judge judge-summ-pick-v1
cnn_mistral_s_n200_v1 · judged/judge-summ-pick-v10.008198judge judge-summ-pick-v1
cnn_deepseek_l_n200_v1 · judged/judge-summ-pick-v10.007198judge judge-summ-pick-v1
cnn_deepseek_s_n200_v1 · judged/judge-summ-pick-v10.007198judge judge-summ-pick-v1
cnn_gemma_m_n200_v1 · judged/judge-summ-pick-v10.006198judge judge-summ-pick-v1
cnn_openai_s_n200_v1 · judged/judge-summ-pick-v10.005198judge judge-summ-pick-v1
cnn_openai_l_n200_v1 · judged/judge-summ-pick-v10.004198judge judge-summ-pick-v1
cnn_gemma_s_n200_v1 · judged/judge-summ-pick-v10.003198judge judge-summ-pick-v1
cnn_llama_s_n200_v1 · judged/judge-summ-pick-v10.003198judge judge-summ-pick-v1
cnn_qwen_s_n200_v1 · judged/judge-summ-pick-v10.002198judge judge-summ-pick-v1
cnn_glm_s_n200_v1 · judged/judge-summ-pick-v10.002198judge judge-summ-pick-v1
cnn_openai_m_n200_v1 · judged/judge-summ-pick-v10.002198judge judge-summ-pick-v1
cnn_gemma_l_n200_v1 · judged/judge-summ-pick-v10.001198judge judge-summ-pick-v1
cnn_llama_m_n200_v1 · judged/judge-summ-pick-v10.001198judge judge-summ-pick-v1
cnn_llama_l_n200_v1 · judged/judge-summ-pick-v10.001198judge judge-summ-pick-v1
cnn_bart_l_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1
cnn_bart_l_xsum_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1
cnn_bart_m_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1
cnn_bart_s_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1
cnn_lead3_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1
cnn_mistral_l_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1
cnn_mistral_m_n200_v1 · judged/judge-summ-pick-v10.000198judge judge-summ-pick-v1