© 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. 509
    Prev
    Next
    INDEX
    Explanations

    section headers like "1. Introduction"

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

    The neuron strongly activates on numeric tokens (standalone digits or numbers) in technical/code-like contexts.

    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
    を行いました
    -2.27
    をしました
    -2.27
    赑
    -2.27
     budu
    -2.22
    所で
    -2.16
     karena
    -2.13
    に来て
    -2.11
     sufrió
    -1.98
    日本的
    -1.98
     cayó
    -1.98
    POSITIVE LOGITS
    0
    3.08
    a
    2.98
     —
    2.59
    an
    2.58
    e
    2.47
    er
    2.38
    ”
    2.33
    ان
    2.13
     .
    2.11
    <sup>
    2.03
    Activations Density 0.017%

    No Known Activations