© 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. 10-GEMMASCOPE-RES-131K
    7. 130295
    Prev
    Next
    INDEX
    Explanations

    cannot get or start

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

    This neuron activates strongly on key action verbs in “can’t” or “cannot” user questions (e.g. get, start, select, work, read, tell, write).

    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
     内衣
    -2.63
    
    -2.55
     花纹
    -2.47
     唯美
    -2.44
     espé
    -2.34
    縠
    -2.30
     rafra
    -2.30
     marte
    -2.28
     ACTUALLY
    -2.28
     douane
    -2.27
    POSITIVE LOGITS
     on
    2.25
    an
    2.09
     de
    2.09
     if
    2.09
    i
    2.02
    T
    1.92
     for
    1.90
     and
    1.90
    C
    1.88
    S
    1.88
    Activations Density 0.038%

    No Known Activations