© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 82752
    Prev
    Next
    INDEX
    Explanations

    android:id attributes

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

    This neuron lights up on code‐style tokens—especially markup attributes (like “android:…”) and CSS/vendor prefixes (like “-webkit”).

    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
     their
    -1.15
     tailored
    -0.99
    meja
    -0.91
     any
    -0.90
     one
    -0.89
     by
    -0.89
     domésticos
    -0.88
    Although
    -0.88
     traduction
    -0.88
    紘
    -0.86
    POSITIVE LOGITS
    −/−
    0.99
     ayur
    0.97
    对他
    0.97
     delightfully
    0.97
     oscuros
    0.94
     descri
    0.91
     intermin
    0.91
     overly
    0.91
     inconsist
    0.90
     тему
    0.90
    Activations Density 0.000%

    No Known Activations