© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 128453
    Prev
    Next
    INDEX
    Explanations

    avoiding negative outcomes or specific actions

    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_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
    ވަ
    0.47
    athon
    0.45
    ՝
    0.43
     दुसऱ्या
    0.43
     نہیں۔
    0.42
     числі
    0.42
    อาจ
    0.42
    약을
    0.41
    मध्यप्रदेश
    0.41
    पणा
    0.41
    POSITIVE LOGITS
     homogen
    0.53
     Insgesamt
    0.45
    全体
    0.45
    Overall
    0.44
     Overall
    0.43
    各
    0.41
    estens
    0.40
    ከት
    0.40
    整體
    0.40
     goblet
    0.39
    Activations Density 0.003%

    No Known Activations