© 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. 19131
    Prev
    Next
    INDEX
    Explanations

    various languages and contexts

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

    The neuron fires on single‐character “s” tokens (including lookalike characters in other scripts, e.g. Latin, Thai, Tibetan, Cyrillic) rather than on full words.

    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
    s
    -3.34
    ↵
    -2.75
     praktisch
    -2.59
    Ꮈ
    -2.55
    us
    -2.53
     нач
    -2.52
    )—
    -2.45
    总是
    -2.44
     verdient
    -2.44
     nimmer
    -2.42
    POSITIVE LOGITS
     is
    3.13
    borderSide
    2.70
    ).
    2.69
    谕
    2.52
     К
    2.50
    闥
    2.48
    providedIn
    2.48
    rierten
    2.44
    辄
    2.44
    PopMatrix
    2.38
    Activations Density 0.000%

    No Known Activations