© 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-65K
    7. 7213
    Prev
    Next
    INDEX
    Explanations

    unauthorized, unknown, never

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

    words indicating refusal, warnings, or ethical objections to harmful requests.

    oai_token-act-pair · claude-4-5-haikuTriggered by @jamesnaruto04
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_65k_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
     déprim
    0.44
    லா
    0.43
    VisualStyle
    0.41
     кое
    0.41
     depressing
    0.41
     قابل
    0.40
    难以
    0.40
     करताना
    0.40
    可用
    0.40
    អាច
    0.40
    POSITIVE LOGITS
    知らない
    0.60
     अनजान
    0.59
     unauthorized
    0.56
     unfounded
    0.56
     inexist
    0.56
     unknowingly
    0.56
     never
    0.55
    不存在
    0.55
     nunca
    0.55
     desconoc
    0.55
    Activations Density 0.154%

    No Known Activations