© 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. 113144
    Prev
    Next
    INDEX
    Explanations

    with feelings or mannerisms

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

    The neuron fires on instances of the preposition “with” used to introduce a manner or attitude phrase immediately following a verb (e.g. “asked with…,” “treated with…,” “watched with…,” “looked on with interest,” etc.).

    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
     of
    -2.19
     ABBRE
    -1.56
    rozco
    -1.51
     istimewa
    -1.50
    N
    -1.48
    bosity
    -1.48
    を用
    -1.43
     DEPOSITORY
    -1.40
    ymas
    -1.38
    R
    -1.34
    POSITIVE LOGITS
     Mungkin
    1.41
    $\$
    1.39
    极为
    1.38
     komplette
    1.38
     halbe
    1.35
    十分的
    1.34
     weiße
    1.32
     saludables
    1.31
     cansado
    1.31
     embarazada
    1.30
    Activations Density 0.037%

    No Known Activations