Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-3-27B-IT
    3. 57-GEMMASCOPE-2-TRANSCODER-262K
    4. 229039
    Prev
    Next
    INDEX
    Explanations

    context specific

    np_max-act-logits · gemini-2.5-flash-lite

    prominent content nouns—especially in Korean (and sometimes other non-English text)—that denote key entities, roles, or topics.

    oai_token-act-pair · gpt-5Triggered by @yooniel31
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_57_width_262k_l0_small_affine
    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
     leveling
    0.38
    在中
    0.38
    uenza
    0.36
     टेंपरे
    0.36
     Dene
    0.36
    ରେ
    0.36
     Aire
    0.35
     repe
    0.35
    hape
    0.34
    insertOne
    0.34
    POSITIVE LOGITS
    답
    0.45
     답
    0.40
    라는
    0.39
    축
    0.39
     что
    0.38
     라는
    0.38
    oure
    0.37
     мог
    0.37
     violently
    0.36
     Plea
    0.36
    Activations Density 0.000%

    No Known Activations