© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 1943
    Prev
    Next
    INDEX
    Explanations

    assaults and illegal actions

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

    The neuron detects long, structured assistant responses that provide detailed, step‑by‑step procedural or safety/advisory guidance.

    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
     fejleszt
    0.45
     оптими
    0.44
    优雅
    0.44
    纳入
    0.43
     Optimization
    0.43
     زمانے
    0.43
     nhỉ
    0.42
     বিশ্লেষণের
    0.42
     বিখ্যাত
    0.42
     Depreciation
    0.42
    POSITIVE LOGITS
     altercation
    0.82
     clothing
    0.65
     unlawfully
    0.62
     witnesses
    0.60
     वारदात
    0.60
     assailant
    0.60
     intoxicated
    0.59
     incapacitated
    0.58
     commotion
    0.58
     concealed
    0.57
    Activations Density 0.042%

    No Known Activations