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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-4B-IT
    5. 17-GEMMASCOPE-2-RES-16K
    7. 5657
    Prev
    Next
    INDEX
    Explanations

    acknowledging or admitting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_17_width_16k_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
    orce
    1.73
    د
    1.71
     нес
    1.69
     дела
    1.66
    ट
    1.65
    اب
    1.58
     elaborate
    1.57
     densely
    1.54
    cane
    1.53
    联
    1.50
    POSITIVE LOGITS
     defeat
    2.58
    defeat
    2.10
    acknowledge
    1.88
    Acknowledgment
    1.86
     defeats
    1.86
     paternity
    1.86
    InputCommand
    1.85
    ना
    1.83
     acknowledge
    1.78
     ack
    1.74
    Activations Density 0.265%

    No Known Activations