© 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. 25784
    Prev
    Next
    INDEX
    Explanations

    sensible and rational approach

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

    The neuron fires on words that denote cognitive perception or reasoning (e.g. see, sense, realise, look, think, calm, scientific).

    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
    اہ
    -0.90
    极致
    -0.85
    Unfinished
    -0.82
     åren
    -0.82
     τέ
    -0.81
    🥘
    -0.81
     couvert
    -0.81
    มาณ
    -0.81
     Comentarios
    -0.81
    ಿಯ
    -0.80
    POSITIVE LOGITS
     rational
    1.36
    冷静
    1.31
     calmly
    1.23
     rationally
    1.23
     sober
    1.09
     calm
    1.09
     logical
    1.09
     calmed
    1.07
    rational
    1.04
     calmer
    1.04
    Activations Density 0.022%

    No Known Activations