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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 12025
    Prev
    Next
    INDEX
    Explanations

    deal

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

    The neuron detects salient content words—especially nouns and named entities that mark the main topic or focal concepts in a passage.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    𝜌
    0.42
    。『
    0.39
     entweder
    0.39
     Andere
    0.39
     क्रमशः
    0.39
    !!!!!!!!!!!!!!!!
    0.39
    """"
    0.38
    🦗
    0.38
    ===========
    0.38
     Beide
    0.38
    POSITIVE LOGITS
     ‘
    0.42
    <strong>
    0.40
     overall
    0.39
     was
    0.37
     Overall
    0.36
    hg
    0.36
     Festival
    0.36
     begun
    0.35
     ${
    0.35
     $\
    0.34
    Activations Density 0.054%

    No Known Activations