© 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. 59345
    Prev
    Next
    INDEX
    Explanations

    bar

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

    This neuron primarily activates on occurrences of the token “bar.”

    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
    1
    -2.84
    
    -2.44
    The
    -2.31
    ↵
    -2.28
    4
    -2.25
     VERY
    -2.19
    很不
    -2.16
     Then
    -2.11
    、“
    -2.08
                    
    -2.06
    POSITIVE LOGITS
    bar
    3.14
    ’;
    2.73
    ⃚
    2.48
    u
    2.47
    餮
    2.39
    鹛
    2.38
    苹
    2.34
     возникает
    2.30
    čiau
    2.27
    螄
    2.27
    Activations Density 0.007%

    No Known Activations