© 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. 33423
    Prev
    Next
    INDEX
    Explanations

    of the main ingredient

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

    The neuron fires on the heads of multi‐word compound noun phrases (e.g. technical or descriptive noun–noun sequences like “rotor blade,” “system icon,” “injecting drug user,” 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_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
     universelle
    -1.17
    ducers
    -1.16
    沨
    -1.12
     intér
    -1.11
     bakter
    -1.08
     épis
    -1.05
     sympathique
    -1.04
    sudadera
    -1.04
     kredi
    -1.04
     induk
    -1.02
    POSITIVE LOGITS
     either
    1.45
     from
    1.23
     these
    1.16
     its
    1.13
    這些
    1.11
     insight
    1.07
     These
    0.98
     aceste
    0.91
     or
    0.91
     whose
    0.90
    Activations Density 0.151%

    No Known Activations