© 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. 10-GEMMASCOPE-RES-131K
    7. 103937
    Prev
    Next
    INDEX
    Explanations

    housing regulatory arts barrier Freder factor Research Inbox

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

    The neuron lights up on individual tokens that are part of multiword names or technical compound terms (e.g. proper nouns, hyphenated or specialized phrases).

    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
    
    -1.40
     your
    -1.36
    糶
    -1.30
     instead
    -1.29
     speciale
    -1.28
     ziemi
    -1.28
     maniere
    -1.27
     vitale
    -1.27
     vostre
    -1.26
     materie
    -1.26
    POSITIVE LOGITS
     including
    1.48
     new
    1.47
    ;
    1.38
     those
    1.37
     wholly
    1.36
    setTypeface
    1.35
     such
    1.34
     includes
    1.33
    فيلم
    1.33
     may
    1.32
    Activations Density 0.231%

    No Known Activations