© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 15318
    Prev
    Next
    INDEX
    Explanations

    pronoun + past tense verb

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

    The neuron strongly activates on third‐person male pronouns (especially “He” and “His”) in biographical narrative sentences.

    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
    iously
    -1.43
    atized
    -1.40
    iennes
    -1.25
     garantire
    -1.23
    ulates
    -1.23
     offrire
    -1.22
    ☗
    -1.20
    ligencia
    -1.20
    atively
    -1.19
    paramInt
    -1.16
    POSITIVE LOGITS
     first
    1.72
     of
    1.66
     from
    1.56
     również
    1.47
     only
    1.44
     after
    1.41
     then
    1.39
     at
    1.34
     through
    1.32
     when
    1.26
    Activations Density 0.151%

    No Known Activations