© 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. 27296
    Prev
    Next
    INDEX
    Explanations

    existing data or identifiers

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

    This neuron fires on code identifiers and XML/HTML tag names—i.e. uppercase or camel-case tokens used as element names, parameters, or function/class identifiers.

    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
     päät
    -0.90
    pooling
    -0.89
     배우
    -0.88
    création
    -0.88
     kaikk
    -0.88
     hänen
    -0.85
     şun
    -0.85
     hediye
    -0.84
    ejercicio
    -0.82
     нового
    -0.82
    POSITIVE LOGITS
     previous
    1.13
     existing
    1.13
     old
    1.10
    ก่อน
    1.06
     previously
    1.02
     Besøkt
    1.02
    เดิม
    1.00
    ToEdit
    0.97
    щал
    0.97
    ToUpdate
    0.96
    Activations Density 0.004%

    No Known Activations