© 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-262K
    7. 123566
    Prev
    Next
    INDEX
    Explanations

    formal closing or tries

    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_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
    atur
    0.51
    ield
    0.47
    acters
    0.45
    蛳
    0.45
    nate
    0.44
    ional
    0.44
     Packaging
    0.43
    inary
    0.42
    ructuring
    0.42
    
    0.42
    POSITIVE LOGITS
     recop
    0.43
     convuls
    0.43
     pretends
    0.43
     plu
    0.43
     visc
    0.42
     impide
    0.42
     embarazo
    0.42
     cơn
    0.41
     inexplic
    0.40
    विदा
    0.40
    Activations Density 0.002%

    No Known Activations