© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 10838
    Prev
    Next
    INDEX
    Explanations

    enter or run

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

    Cyrillic verb forms, particularly infinitives and conjugated verbs ending in specific suffixes.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on Cyrillic verbs—especially present‐tense or third‐person verb forms in Russian.

    oai_token-act-pair · o4-miniTriggered 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
    訁
    -1.98
    that
    -1.84
    titial
    -1.80
    鸸
    -1.74
    𝅥
    -1.74
    それを
    -1.73
    after
    -1.72
    鸲
    -1.65
    あっという
    -1.64
     you
    -1.63
    POSITIVE LOGITS
     in
    2.52
     on
    2.05
    ;
    1.84
     These
    1.80
     by
    1.80
     like
    1.75
     összefoglal
    1.64
     Such
    1.63
    }
    1.58
    ?'
    1.56
    Activations Density 0.020%

    No Known Activations