© 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. 47527
    Prev
    Next
    INDEX
    Explanations

    AS IS BASIS

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

    The neuron selectively fires on relatively rare, multi-syllable content words (i.e. topic-specific keywords) rather than common function words.

    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
     our
    -1.72
    ...</
    -1.58
    ruguay
    -1.47
    (!__
    -1.45
     it
    -1.40
    )\\
    -1.35
    ētu
    -1.31
    ẹo
    -1.30
     polémica
    -1.29
     about
    -1.28
    POSITIVE LOGITS
     is
    1.54
    There
    1.51
    Its
    1.41
    If
    1.38
     koska
    1.33
    Why
    1.31
    They
    1.30
    DAFTAR
    1.28
    Whether
    1.28
    Pře
    1.27
    Activations Density 0.194%

    No Known Activations