© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73190
    Prev
    Next
    INDEX
    Explanations

    legal citations

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

    The neuron strongly responds to the presence of parentheses—i.e. ‘(’ and ‘)’—as they appear in function/method signatures, type casts, switch statements, arrow functions, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    salomon
    -1.09
     those
    -1.05
    maksud
    -1.01
     also
    -1.01
    bundes
    -1.00
    tvguide
    -0.99
     Kleid
    -0.99
     czter
    -0.98
    motorola
    -0.98
    Those
    -0.97
    POSITIVE LOGITS
     })}
    1.04
     usati
    1.00
     integrale
    1.00
    }}
    0.98
    vgl
    0.98
     INTERFACE
    0.98
    ¥
    0.96
     }))
    0.95
    toString
    0.95
    })));
    0.94
    Activations Density 0.001%

    No Known Activations