© 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. 3093
    Prev
    Next
    INDEX
    Explanations

    This neuron detects occurrences of the verb “convert” (in its various forms and tenses).

    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
    馳
    0.64
     recicl
    0.63
    arker
    0.61
     улучшения
    0.60
     devlet
    0.57
    窃
    0.56
    投资
    0.55
    ্রিয়া
    0.55
    塚
    0.55
     steers
    0.55
    POSITIVE LOGITS
     Convert
    0.93
     into
    0.89
    Convert
    0.81
    转换为
    0.79
    into
    0.78
     convert
    0.76
     Into
    0.76
    convert
    0.74
    Converting
    0.73
     Converting
    0.71
    Activations Density 0.025%

    No Known Activations