© 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. 34-GEMMASCOPE-RES-131K
    7. 57087
    Prev
    Next
    INDEX
    Explanations

    blank or empty

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

    The neuron fires strongly on words and phrases in form‐validation error messages (e.g. “blank,” “missing,” “empty”).

    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
     Eligible
    -0.76
    čů
    -0.76
    winding
    -0.73
    ensuche
    -0.71
    κη
    -0.71
    溺
    -0.71
    Eligible
    -0.71
    bergs
    -0.69
    ligt
    -0.68
    uating
    -0.68
    POSITIVE LOGITS
     blank
    5.44
    blank
    4.47
     Blank
    3.92
    Blank
    3.92
    BLANK
    3.48
     empty
    3.41
     blanks
    3.33
    空白
    3.28
     BLANK
    3.17
    empty
    2.72
    Activations Density 0.085%

    No Known Activations