© 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. 13013
    Prev
    Next
    INDEX
    Explanations

    Samsung product announcements

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

    The neuron activates on standalone numeric tokens (sequences of digits).

    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
     after
    -1.33
     espagnol
    -1.17
     when
    -1.16
     ilustracji
    -1.11
    رایی
    -1.09
     will
    -1.09
    erdings
    -1.08
    した後
    -1.06
     belge
    -1.05
    Lett
    -1.05
    POSITIVE LOGITS
    ↵↵↵↵↵↵↵↵↵↵↵
    1.20
     focuses
    1.18
     agregó
    1.16
     explicó
    1.15
     aseguró
    1.13
     afirmó
    1.12
    <h3>
    1.12
     reempla
    1.11
    ↵↵
    1.07
    ↵↵↵
    1.06
    Activations Density 0.020%

    No Known Activations