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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 104626
    Prev
    Next
    INDEX
    Explanations

    related questions links media

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

    The neuron detects the “Related” section headings (e.g. “Related Posts,” “Related Questions,” etc.) that mark related‐content links.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     donnent
    -1.38
    ernacle
    -1.34
    іга
    -1.32
    による
    -1.31
    nical
    -1.31
    翆
    -1.30
     Anreg
    -1.30
    ochond
    -1.30
     cotidiano
    -1.27
    がよく
    -1.25
    POSITIVE LOGITS
     and
    1.42
     setelan
    1.38
    ・「
    1.37
     showcasing
    1.36
    </em>
    1.34
     seventeen
    1.34
    cm
    1.34
     things
    1.33
     pushes
    1.32
     fifteen
    1.31
    Activations Density 0.020%

    No Known Activations