© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 66338
    Prev
    Next
    INDEX
    Explanations

    intensifier + strong adjective

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

    The neuron fires on adverbs and modifiers that heighten or intensify meaning (e.g. “even,” “overtly,” “downright,” “bordering on,” “not”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    不會
    -1.00
     horrend
    -0.98
     sooo
    -0.96
     bukanlah
    -0.96
     soooo
    -0.95
    ETTI
    -0.94
    Kilde
    -0.93
    絕對
    -0.93
    這麼
    -0.93
     horri
    -0.93
    POSITIVE LOGITS
     downright
    2.11
     outright
    1.89
     sogar
    1.46
     chiar
    1.23
     almost
    1.14
     addirittura
    1.09
     casi
    1.08
     jopa
    1.05
     in
    1.04
     at
    1.04
    Activations Density 0.069%

    No Known Activations