© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 99965
    Prev
    Next
    INDEX
    Explanations

    project identifiers

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

    The neuron activates on all‐caps tokens or acronyms (e.g. MET, FLAG, APR-246) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     benevolent
    -2.77
    鐐
    -2.66
    翃
    -2.59
    剃
    -2.58
     cuando
    -2.58
    ſelves
    -2.58
    ations
    -2.45
     engage
    -2.39
    籥
    -2.36
     \&
    -2.34
    POSITIVE LOGITS
    蚡
    3.17
    咴
    3.16
    舠
    3.14
    AdvEx
    3.08
     that
    3.06
    
    3.00
    埽
    3.00
    赑
    2.98
    鹮
    2.97
    鱨
    2.92
    Activations Density 0.005%

    No Known Activations