© 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. 34-GEMMASCOPE-RES-131K
    7. 106806
    Prev
    Next
    INDEX
    Explanations

    MHV, MH, mch abbreviations

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

    document codes and file references with "MH" or markup/programming language elements.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron specifically detects LaTeX package or environment names in code (e.g. “mdframed,” “math,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     parezca
    -0.91
     poffible
    -0.86
     reafon
    -0.86
     without
    -0.84
     londres
    -0.83
    чье
    -0.82
     Wallpapers
    -0.81
    IEW
    -0.80
     occafion
    -0.80
    ósz
    -0.80
    POSITIVE LOGITS
     SEEN
    0.88
     CET
    0.88
    vantaged
    0.86
    臜
    0.82
    Competitive
    0.81
     battre
    0.81
    もしっかり
    0.81
    whenever
    0.81
    になりますが
    0.80
    sistence
    0.79
    Activations Density 0.003%

    No Known Activations