© 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. 71318
    Prev
    Next
    INDEX
    Explanations

    explaining the reasoning behind

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

    This neuron activates on the word “behind” when it’s used to introduce or discuss an underlying reason, purpose, or motivation.

    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
    )
    -2.86
     S
    -2.34
     more
    -2.14
     When
    -2.06
     p
    -2.05
     singled
    -2.05
     up
    -2.03
     g
    -2.02
     for
    -2.02
     plethora
    -2.00
    POSITIVE LOGITS
     pérd
    2.69
    隳
    2.38
     reager
    2.36
    岨
    2.23
    蔫
    2.22
    
    2.19
    蠃
    2.17
     them
    2.11
    ⠡
    2.11
    硨
    2.09
    Activations Density 0.010%

    No Known Activations