© 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. 2843
    Prev
    Next
    INDEX
    Explanations

    next token, next sequence

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

    The neuron fires most strongly on object-method chaining syntax in code—i.e. sequences of “.” calls like `.parent().next()` rather than plain text.

    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
    られました
    -0.94
     récemment
    -0.90
    最後に
    -0.90
    removeAll
    -0.89
    Ahol
    -0.87
     recente
    -0.86
     동안
    -0.85
    bitos
    -0.84
    só
    -0.84
    НАЯ
    -0.83
    POSITIVE LOGITS
     next
    4.09
     getNext
    2.80
     다음
    2.77
    next
    2.70
    Next
    2.66
     следу
    2.64
    getNext
    2.58
     następ
    2.55
     Next
    2.53
     NEXT
    2.38
    Activations Density 0.047%

    No Known Activations