© 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. 12848
    Prev
    Next
    INDEX
    Explanations

    Parallel lines or processing

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

    This neuron fires on words that convey directness or straightforwardness in the text.

    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
    arning
    0.70
    юк
    0.64
    트롤
    0.63
     অর্ধ
    0.63
    蜍
    0.62
    क्षण
    0.61
    cych
    0.60
    ilas
    0.58
    襄
    0.58
    ingly
    0.57
    POSITIVE LOGITS
     straight
    1.09
    straight
    1.09
     STRAIGHT
    1.01
     Straight
    0.98
     strait
    0.91
    Straight
    0.89
     straighten
    0.85
     straightened
    0.82
     직선
    0.80
    jacket
    0.67
    Activations Density 0.148%

    No Known Activations