© 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. 22-GEMMASCOPE-RES-131K
    7. 26159
    Prev
    Next
    INDEX
    Explanations

    its condition or state

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

    The neuron responds to very common short “function” words (e.g. of, to, is, we, it, what) that serve as grammatical glue rather than content.

    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
     and
    -1.19
     muy
    -1.01
     worin
    -0.98
     gracias
    -0.98
     numberWith
    -0.94
    versial
    -0.93
    に関しては
    -0.91
     considera
    -0.91
    이나
    -0.89
    ¨
    -0.89
    POSITIVE LOGITS
     its
    3.83
     it
    2.52
    它
    2.14
    它的
    2.11
    Its
    2.00
     Its
    1.94
     его
    1.81
    将其
    1.80
     它
    1.80
     ее
    1.79
    Activations Density 0.113%

    No Known Activations