© 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. 73030
    Prev
    Next
    INDEX
    Explanations

    words in quotes

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

    The neuron fires on words or phrases that are set off by quotation marks or emphasis markers (e.g., surrounded by “ ”, ’ ’, or * *).

    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
     Atención
    -1.52
    ⟣
    -1.48
    乄
    -1.45
     Inoltre
    -1.45
     poème
    -1.38
     бампер
    -1.38
    了过去
    -1.36
     simpl
    -1.34
     neces
    -1.34
    响起
    -1.32
    POSITIVE LOGITS
     variadas
    1.52
     individuals
    1.51
     methods
    1.47
     activities
    1.46
     how
    1.46
     looking
    1.45
     versions
    1.43
     ones
    1.41
     ways
    1.41
     individual
    1.39
    Activations Density 0.042%

    No Known Activations