© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 118997
    Prev
    Next
    INDEX
    Explanations

    cutting off body parts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     yık
    -0.75
     détruire
    -0.75
     حداقل
    -0.74
    excel
    -0.73
    drav
    -0.71
    ㄆ
    -0.71
    cognito
    -0.71
     concentración
    -0.69
     اكتوبر
    -0.69
    clyde
    -0.68
    POSITIVE LOGITS
     severed
    2.38
     chopped
    1.92
     decap
    1.86
     chop
    1.70
     sliced
    1.66
     chopping
    1.64
     headless
    1.63
     cut
    1.57
     detached
    1.46
    decap
    1.45
    Activations Density 0.034%

    No Known Activations