© 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. 113420
    Prev
    Next
    INDEX
    Explanations

    programming language data structures

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

    commas that separate type parameters in generic declarations (like `Map<String, Object>`).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates on numeric literal tokens (i.e. numbers) in code.

    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
    1
    -2.33
    .
    -2.17
    If
    -2.16
    should
    -1.80
    3
    -1.73
    鶿
    -1.70
    自己是
    -1.65
    はじめて
    -1.63
    下意识
    -1.59
    也沒有
    -1.59
    POSITIVE LOGITS
     this
    2.06
     that
    2.03
    <td>
    2.02
    BITDA
    1.91
     become
    1.85
     a
    1.80
     including
    1.80
    erintah
    1.79
     இந்த
    1.76
     Songtext
    1.75
    Activations Density 0.002%

    No Known Activations