© 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-65K
    7. 787
    Prev
    Next
    INDEX
    Explanations

    harmful deceptive malicious

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_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
     kilomètres
    0.45
     périphérique
    0.45
     अधिसूचना
    0.41
     arrang
    0.41
     abstra
    0.41
     fotograf
    0.40
     coucher
    0.40
    汥
    0.40
    غيرة
    0.40
     ovoj
    0.40
    POSITIVE LOGITS
     harmful
    1.34
    有害
    1.09
     harm
    1.02
     dangerous
    0.99
     Harm
    0.99
    harm
    0.98
     harms
    0.96
     harming
    0.91
     berbahaya
    0.91
     malicious
    0.90
    Activations Density 1.020%

    No Known Activations