© 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-IT
    3. 31-GEMMASCOPE-2-RES-65K
    4. 53145
    Prev
    Next
    INDEX
    Explanations

    not(Note: The pattern "not" followed by specific contextual words appears frequently, e.g., "not a version of ChatGPT", "not effective", and the other lists hint at various specific contexts like biology/medical, professional settings, and operations, making "not" a recurring element suggesting exceptions or negations.)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_65k_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
    ઔ
    0.48
    psin
    0.44
    voxel
    0.43
    Probit
    0.43
    perceptron
    0.43
     Cochran
    0.42
     어떻게
    0.42
    🍚
    0.42
    zinha
    0.42
    лили
    0.42
    POSITIVE LOGITS
     Operation
    0.46
     Stop
    0.44
     Elev
    0.44
     operation
    0.43
     Start
    0.43
     Ensure
    0.43
     Inst
    0.42
     All
    0.42
     Safety
    0.41
    iq
    0.41
    Activations Density 0.014%

    No Known Activations