© 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-3-270M-IT
    5. 12-GEMMASCOPE-2-RES-65K
    7. 1766
    Prev
    Next
    INDEX
    Explanations

    harmful or hateful content descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-270m-it/resid_post/layer_12_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
     stal
    1.22
     completes
    1.18
     secures
    1.15
    Solve
    1.12
     cinch
    1.09
    囝
    1.08
     retirement
    1.07
     laurels
    1.06
    蠔
    1.06
     reunion
    1.05
    POSITIVE LOGITS
     hateful
    3.20
     harmful
    3.11
     odio
    2.75
     hatred
    2.73
     derogatory
    2.71
    恶意
    2.67
     slander
    2.66
     оско
    2.65
     malicious
    2.64
     maliciously
    2.63
    Activations Density 3.667%

    No Known Activations