© 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. 66263
    Prev
    Next
    INDEX
    Explanations

    updating or altering information

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

    This neuron detects imperative step verbs in procedural instructions (e.g. “Convert,” “Update,” etc.).

    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
    trying
    -1.51
     papiers
    -1.49
    perhaps
    -1.47
    providing
    -1.45
    losing
    -1.44
    despite
    -1.41
    gives
    -1.37
    discussed
    -1.37
    needed
    -1.34
    having
    -1.34
    POSITIVE LOGITS
     the
    2.20
     your
    1.98
     their
    1.95
     our
    1.78
     with
    1.76
     all
    1.68
     each
    1.67
     every
    1.64
     EVERY
    1.56
     one
    1.55
    Activations Density 0.023%

    No Known Activations