© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 99852
    Prev
    Next
    INDEX
    Explanations

    Cumberland, SCP, pussy

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on capitalized named entities and technical acronyms (e.g. “Cumberland,” “SCP,” “EDC”), signaling recognition of proper names and all-caps abbreviations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ’
    -3.22
    鵟
    -2.75
    .''
    -2.70
    毉
    -2.66
     SimpleName
    -2.61
    錵
    -2.59
    r
    -2.58
    苼
    -2.56
    螄
    -2.55
    ...?
    -2.52
    POSITIVE LOGITS
    艫
    3.05
     связан
    2.39
    缡
    2.38
    ↵↵
    2.36
    僰
    2.33
    絝
    2.31
    および
    2.30
     WITH
    2.27
    ti
    2.27
    })
    2.25
    Activations Density 0.050%

    No Known Activations