© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6056
    Prev
    Next
    INDEX
    Explanations

    `type="`

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

    The neuron strongly activates on common clause-linking subwords and hyphenated or apostrophed fragments (e.g. “in order,” “along,” “shouldn’t,” “it’s,” “well-,” “-biotic”), essentially detecting discourse-connector and transition pieces.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    }\
    0.97
    }[
    0.96
    }
    0.88
    }}\
    0.79
    }~
    0.77
    }\\
    0.76
    }~\
    0.75
    }-\
    0.73
    ~\
    0.71
    }}[
    0.71
    POSITIVE LOGITS
     ruthless
    0.65
     varier
    0.65
     pokaz
    0.64
     suitors
    0.64
     strangers
    0.63
     restless
    0.63
     overdue
    0.62
     merciless
    0.61
     sucedió
    0.60
     patented
    0.59
    Activations Density 0.003%

    No Known Activations