© 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. 33378
    Prev
    Next
    INDEX
    Explanations

    verbs indicating relationship or action

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

    The neuron fires on formal scientific/process-oriented verbs (e.g. “interacts,” “functions,” “binds,” “is used,” “is related to”) that describe how system components operate or relate.

    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
     it
    -2.16
     after
    -1.52
     this
    -1.49
     before
    -1.43
     their
    -1.43
     those
    -1.35
     these
    -1.33
     there
    -1.32
     just
    -1.30
     Even
    -1.23
    POSITIVE LOGITS
     kuitenkin
    1.36
     inoltre
    1.36
    particularly
    1.23
    compañ
    1.21
     però
    1.20
     áng
    1.16
     ingrédients
    1.14
     まる
    1.13
     också
    1.12
    ρή
    1.11
    Activations Density 0.202%

    No Known Activations