© 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. 12344
    Prev
    Next
    INDEX
    Explanations

    code delimiters and structure

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

    This neuron detects Romanian-language (non-English) word segments.

    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
    δει
    0.64
     mandar
    0.63
    ezas
    0.63
    osť
    0.63
    если
    0.62
    霈
    0.62
     किसे
    0.61
    olischen
    0.61
    postion
    0.60
    利的
    0.59
    POSITIVE LOGITS
    ↵↵
    1.13
    ↵
    0.76
     aceea
    0.56
     within
    0.55
     understand
    0.51
     Understand
    0.51
     characterise
    0.50
     innerhalb
    0.49
    RIBUN
    0.49
     HOWEVER
    0.49
    Activations Density 0.007%

    No Known Activations