© 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. 10-GEMMASCOPE-RES-131K
    7. 113690
    Prev
    Next
    INDEX
    Explanations

    directed communication

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

    This neuron activates on occurrences of the phrase “talk to” (including variants like “talked to” or “talk to us”).

    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
    ↵
    -2.08
    1
    -1.74
     sayur
    -1.58
     –
    -1.56
     traditionelle
    -1.52
    hrens
    -1.47
    </b>
    -1.45
     jagung
    -1.41
     professionelle
    -1.41
    的に
    -1.38
    POSITIVE LOGITS
     about
    2.38
    犼
    1.89
    \%,
    1.81
    뺏
    1.72
    岿
    1.68
    溘
    1.66
    ፒ
    1.65
    ?”,
    1.65
    !”,
    1.64
    🤚
    1.63
    Activations Density 0.031%

    No Known Activations