© 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. 40-GEMMASCOPE-2-RES-262K
    7. 4000
    Prev
    Next
    INDEX
    Explanations

    Absence, belongs, hypothesis, stickers, password, distillation

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

    tokens that are the initial subword pieces of proper nouns or domain-specific terms (i.e., beginnings/prefixes of names and technical words).

    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_40_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
    руют
    0.17
    фи
    0.15
     ಉತ್ಪನ್ನ
    0.14
    輥
    0.14
     ಮಾಡುವ
    0.14
    με
    0.13
     آئینے
    0.13
    voxel
    0.13
    纡
    0.13
    rescence
    0.13
    POSITIVE LOGITS
     (
    0.16
     This
    0.15
     cf
    0.14
     To
    0.14
     details
    0.14
    ",
    0.14
     offent
    0.14
     this
    0.14
     Jon
    0.14
     don
    0.13
    Activations Density 1.611%

    No Known Activations