© 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. 58101
    Prev
    Next
    INDEX
    Explanations

    direction or destination

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

    The neuron fires on the very first significant content token that begins each new text segment (i.e. it marks the opening word of a block).

    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
    CO
    -0.79
    HA
    -0.77
    HAS
    -0.72
    nn
    -0.72
    MAT
    -0.69
    代の
    -0.69
    addGroup
    -0.68
    override
    -0.68
    ミュ
    -0.67
    stdarg
    -0.66
    POSITIVE LOGITS
    への
    2.56
    へ
    1.57
     unto
    1.44
    కి
    1.38
     onto
    1.30
     към
    1.22
    로
    1.20
     kepada
    1.14
    కు
    1.13
     into
    1.12
    Activations Density 0.026%

    No Known Activations