© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 128248
    Prev
    Next
    INDEX
    Explanations

    The neuron is detecting code punctuation, especially the period “.” (and to a lesser extent semicolons) used in member access and statement termination.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    medical and code terms

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

    C# code importing System namespaces.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     to
    -1.84
     these
    -1.24
     by
    -1.20
     on
    -1.18
     tore
    -1.14
    舠
    -1.14
     during
    -1.13
    GRAFÍA
    -1.13
     actually
    -1.10
    ipeline
    -1.05
    POSITIVE LOGITS
    感じで
    1.40
    場所で
    1.30
    えっ
    1.29
    ほら
    1.23
    ことも
    1.20
    ***
    1.19
    あぁ
    1.18
    さぁ
    1.16
    久し
    1.14
    ところに
    1.13
    Activations Density 0.005%

    No Known Activations