© 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
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 25410
    Prev
    Next
    INDEX
    Explanations

    instruction generation

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

    instructions and meta-prompts that tell the model what task to do, how to format the response, or impose constraints (imperative, second-person directives and guidelines).

    oai_token-act-pair · gpt-5Triggered 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
     entier
    0.41
     ortodox
    0.39
     pertandingan
    0.39
    在
    0.38
     ehm
    0.38
     levens
    0.38
     elkaar
    0.38
     haircuts
    0.37
     automating
    0.37
     paintball
    0.37
    POSITIVE LOGITS
     также
    0.56
     또한
    0.55
    yrıca
    0.52
     Ayrıca
    0.52
     Therefore
    0.51
     ასევე
    0.51
     しかし
    0.50
     inoltre
    0.49
     Also
    0.49
     also
    0.49
    Activations Density 0.192%

    No Known Activations