© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38576
    Prev
    Next
    INDEX
    Explanations

    various explanations

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

    The neuron is acting as a learned position counter – it activates proportionally to a token’s absolute position index in the document.

    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
     kuri
    -0.93
     cartes
    -0.89
     Kto
    -0.88
    brak
    -0.86
     kurios
    -0.85
     aceite
    -0.85
    rophore
    -0.85
     ihre
    -0.85
    fetera
    -0.84
    ütün
    -0.83
    POSITIVE LOGITS
     various
    0.96
    nyk
    0.90
     Various
    0.90
     investimento
    0.89
     investment
    0.87
     explanation
    0.85
    各種
    0.83
    ık
    0.83
    Various
    0.82
    ivid
    0.82
    Activations Density 0.001%

    No Known Activations