unsloth

/

cogito-v2-preview-llama-109B-MoE-GGUF

like 8

Follow

Unsloth AI 8.68k

More MMLU Pro results for ggufs are available here

logs

+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| Model                                          | Dataset   | Metric          | Subset           |   Num |   Score | Cat.0   |
+================================================+===========+=================+==================+=======+=========+=========+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | computer science |    10 |  0.6    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | math             |    10 |  0.6    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | chemistry        |    10 |  1      | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | engineering      |    10 |  0.6    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | law              |    10 |  0.7    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | biology          |    10 |  0.9    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | health           |    10 |  0.8    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | physics          |    10 |  1      | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | business         |    10 |  0.8    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | philosophy       |    10 |  0.6    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | economics        |    10 |  0.8    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | other            |    10 |  0.9    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | psychology       |    10 |  0.7    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | history          |    10 |  0.9    | default |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+
| cogito-v2-preview-llama-109B-MoE-UD-IQ2_M.gguf | mmlu_pro  | AverageAccuracy | OVERALL          |   140 |  0.7786 | -       |
+------------------------------------------------+-----------+-----------------+------------------+-------+---------+---------+