© 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. 200882
    Prev
    Next
    INDEX
    Explanations

    model is a large language

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

    detects the model/assistant role token or the start of an assistant/model response.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    Preference
    0.46
    Proto
    0.45
    And
    0.44
     också
    0.44
    Gb
    0.44
     також
    0.44
    Sequence
    0.43
    屽
    0.43
    աջ
    0.43
    也
    0.43
    POSITIVE LOGITS
     eaters
    0.47
     robbers
    0.47
     chemists
    0.45
     bakteri
    0.45
     haters
    0.45
     tourists
    0.45
    網友
    0.44
     advertisers
    0.43
     dieses
    0.43
     netizens
    0.43
    Activations Density 0.031%

    No Known Activations