© 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. 128404
    Prev
    Next
    INDEX
    Explanations

    The neuron lights up on high-frequency function words—especially pronouns and auxiliary/modal verbs.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    they and their

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     itself
    -1.42
    ᚑ
    -1.40
     its
    -1.39
     jego
    -1.27
     was
    -1.27
    WebElementEntity
    -1.16
    它的
    -1.16
     vostra
    -1.16
    ondale
    -1.14
     kendini
    -1.12
    POSITIVE LOGITS
     themselves
    3.69
     their
    2.98
     they
    2.48
     leurs
    2.34
    themselves
    2.22
    Their
    2.02
    their
    1.97
    他们的
    1.92
     leur
    1.87
     ihren
    1.84
    Activations Density 0.130%

    No Known Activations