© 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. 18062
    Prev
    Next
    INDEX
    Explanations

    AI safety guidelines

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

    Tokens that occur in the model/assistant's generated responses — i.e., parts of the assistant's voice, especially safety, explanatory, or conversational reply text.

    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
    emorrh
    0.47
    ന്ത്രാ
    0.43
     पारी
    0.43
    padStart
    0.43
    ictionaries
    0.42
    ServerError
    0.41
     Conversely
    0.40
     palmit
    0.39
    ndrome
    0.38
     sucesivamente
    0.38
    POSITIVE LOGITS
    غير
    0.47
    غیر
    0.45
    釈
    0.45
     Dairy
    0.44
     بود
    0.43
     Tessa
    0.42
     Sugar
    0.42
     ویس
    0.41
    ですが
    0.40
    ﻁ
    0.40
    Activations Density 14.515%

    No Known Activations