© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 3573
    Prev
    Next
    INDEX
    Explanations

    lists and ranges

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

    The neuron is selectively detecting the curly apostrophe character (’) used in contractions.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     dobbiamo
    0.59
     Djokovic
    0.53
     avem
    0.53
     நிச்சயமாக
    0.52
     vyš
    0.52
     вали
    0.50
     ï
    0.50
     जडेजा
    0.49
     対
    0.49
     acyl
    0.49
    POSITIVE LOGITS
    ैव
    0.50
    For
    0.47
    ].
    0.46
    ))]
    0.46
    ভাবে
    0.45
    None
    0.45
    </tr>
    0.45
    Element
    0.44
    ((
    0.44
    undefined
    0.44
    Activations Density 0.070%

    No Known Activations