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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-65K
    7. 5583
    Prev
    Next
    INDEX
    Explanations

    prevents escalation or progression

    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_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
     Antibody
    0.44
     আসামি
    0.41
     Shield
    0.40
     আসার
    0.40
    Reuse
    0.39
    ANT
    0.39
    ="\
    0.39
    软件
    0.38
    =\
    0.38
    を用いて
    0.38
    POSITIVE LOGITS
     worsening
    0.75
     dangerous
    0.72
     опас
    0.71
     escalating
    0.68
     escalate
    0.68
     escalation
    0.66
     danger
    0.66
     worsen
    0.63
     nguy
    0.62
     serious
    0.61
    Activations Density 0.138%

    No Known Activations