© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66316
    Prev
    Next
    INDEX
    Explanations

    in regard to or with reference to

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

    This neuron detects formal linking phrases, especially prepositional constructions like “with regard to…” or “with reference to…”.

    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
    tocks
    -1.61
    phans
    -1.56
    rée
    -1.55
    grains
    -1.48
    melons
    -1.46
     sér
    -1.44
     tuli
    -1.44
     volumen
    -1.43
     sonore
    -1.43
     salles
    -1.38
    POSITIVE LOGITS
     to
    1.69
     огромное
    1.57
     that
    1.52
    '
    1.52
     Können
    1.48
    sucurs
    1.47
     jednotliv
    1.44
    不是
    1.42
     Ähn
    1.42
     possibilité
    1.40
    Activations Density 0.024%

    No Known Activations