© 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. 27417
    Prev
    Next
    INDEX
    Explanations

    so what

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

    The neuron fires strongly on question‐introducing words and phrases (e.g. “what,” “why,” “so…,” “let’s…”)—i.e. markers that launch a question or hypothetical.

    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
    趼
    -1.24
     thofe
    -1.13
    Bara
    -1.09
     akad
    -1.05
    fordshire
    -1.05
    ſſen
    -1.05
     darbo
    -1.01
    licitations
    -0.99
    mô
    -0.98
    驶
    -0.98
    POSITIVE LOGITS
                
    1.08
            
    1.07
    But
    1.00
     enam
    0.99
     当
    0.94
    知名
    0.94
    [
    0.94
     不能
    0.92
     //
    0.91
     今日
    0.91
    Activations Density 0.084%

    No Known Activations