© 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
    2. Gemma-3-27B
    3. 53-GEMMASCOPE-2-RES-65K
    4. 17415
    Prev
    Next
    INDEX
    Explanations

    to lie

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

    lying after underlying

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

    important semantic content words that carry key meaning in a passage.

    oai_token-act-pair · claude-4-5-haikuTriggered by @ananthanv94
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_53_width_65k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    すす
    0.44
    སྐ
    0.37
     ショルダー
    0.37
    UME
    0.36
     भा
    0.35
     hInstance
    0.34
    িজে
    0.33
    锶
    0.33
    シオ
    0.33
    pix
    0.32
    POSITIVE LOGITS
     lie
    4.13
     lies
    4.13
     lying
    4.13
     Lie
    3.69
    Lie
    3.53
     лежа
    3.48
     лежит
    3.48
     LIE
    3.47
     nằm
    3.36
    躺
    3.28
    Activations Density 0.208%

    No Known Activations