© 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. 26386
    Prev
    Next
    INDEX
    Explanations

    "to" followed by a verb

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

    The neuron activates on the “to” token when it’s used as the infinitive‐marker introducing a following verb.

    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
    一脸
    -1.24
    étique
    -1.18
     recientemente
    -1.14
    立马
    -1.14
    ously
    -1.13
    抿
    -1.11
    は
    -1.08
    搁
    -1.08
    を利用
    -1.07
    좆
    -1.06
    POSITIVE LOGITS
    Törté
    1.42
     get
    1.32
    为什么
    1.30
     begin
    1.29
    न
    1.27
    Why
    1.26
     be
    1.23
     They
    1.23
     perform
    1.09
    şey
    1.09
    Activations Density 0.015%

    No Known Activations