© 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. 14697
    Prev
    Next
    INDEX
    Explanations

    outline or lay out plans

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

    The neuron fires strongly on “X out” or related constructions (laid out, set out, outlined, spelled out, etc.) that introduce explanations or step-by-step details.

    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
    2
    -1.84
    {
    -1.76
    \\
    -1.57
    的
    -1.44
    the
    -1.40
     ваших
    -1.38
    =@"
    -1.35
     piste
    -1.35
    .
    -1.34
    ",
    -1.34
    POSITIVE LOGITS
     how
    1.65
     kteří
    1.63
     required
    1.63
     universidades
    1.55
    puestas
    1.55
    !’
    1.53
     lidé
    1.52
     However
    1.51
     guidelines
    1.49
    ophobia
    1.49
    Activations Density 0.025%

    No Known Activations