© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 41410
    Prev
    Next
    INDEX
    Explanations

    force, presence, voice

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

    The neuron fires on nouns that denote a powerful or influential entity—words like “force,” “presence,” “difference-maker,” or “voice” that signal strength or impact.

    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
     muchísimo
    -1.66
    ثم
    -1.55
     facilidad
    -1.52
    ítmény
    -1.50
    daging
    -1.50
    ったので
    -1.46
     perfectamente
    -1.45
     básicamente
    -1.44
     repuesto
    -1.42
     élégante
    -1.42
    POSITIVE LOGITS
     in
    1.98
    }
    1.74
    end
    1.63
    ent
    1.59
     of
    1.56
    くださった
    1.42
    he
    1.41
     typelib
    1.41
    ctive
    1.40
    "];
    
    1.39
    Activations Density 0.046%

    No Known Activations