© 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. 22-GEMMASCOPE-RES-131K
    7. 69197
    Prev
    Next
    INDEX
    Explanations

    financial, geographical, and biological terms

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

    proper nouns or named entities with emphasis on technical or geographic terms.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron selectively activates on three-letter, all-uppercase tokens (e.g., Fin, Ur, Hun, Dire).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ↵↵
    -2.88
    部分は
    -2.39
    <bos>
    -2.33
     
    -2.33
    .
    -2.27
     a
    -2.16
    which
    -2.06
    渽
    -2.06
    時も
    -2.05
    这也
    -2.02
    POSITIVE LOGITS
    埽
    2.55
    鼉
    2.53
    entikan
    2.45
    老闆
    2.41
    odied
    2.28
     humeur
    2.20
    {
    2.19
    쫑
    2.16
    1
    2.16
    ?"
    2.11
    Activations Density 0.161%

    No Known Activations