© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 73018
    Prev
    Next
    INDEX
    Explanations

    свой, мой, ней

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

    The neuron is detecting Cyrillic‐script characters (i.e. Russian/Ukrainian/Serbian letters).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the presence of the Cyrillic “short i” letter (й/ј), especially in word endings like -й, -ий, -ой, -ей across Slavic-language text.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    Cyrillic text, particularly in Russian and Ukrainian languages.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ጺ
    -3.36
    郿
    -2.80
     allow
    -2.67
    ⃢
    -2.58
    鋮
    -2.56
    ቍ
    -2.55
    劻
    -2.55
    骝
    -2.50
     zza
    -2.48
     Coruña
    -2.47
    POSITIVE LOGITS
    й
    2.89
    –
    2.78
            
    2.78
    みると
    2.70
    みたら
    2.61
     VERY
    2.58
     
    2.55
    みよう
    2.52
     Descri
    2.42
    となります
    2.41
    Activations Density 0.004%

    No Known Activations