© 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. 73380
    Prev
    Next
    INDEX
    Explanations

    quoting or reporting text

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

    The neuron detects question phrases asking for definitions or clarifications—especially “what … mean by …”.

    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
    ఝ
    -0.89
    räger
    -0.88
    használ
    -0.88
    णी
    -0.85
    ఛ
    -0.81
    СП
    -0.80
    всем
    -0.80
     colonne
    -0.79
     maig
    -0.78
    cemos
    -0.78
    POSITIVE LOGITS
    らの
    0.96
    ご了承ください
    0.86
     include
    0.81
    安徽
    0.81
     comprises
    0.81
    Natürlich
    0.80
    Detalhes
    0.80
     ",
    0.79
     includes
    0.79
    の中
    0.78
    Activations Density 0.090%

    No Known Activations