© 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. 29188
    Prev
    Next
    INDEX
    Explanations

    reach out to

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

    The neuron fires on occurrences of the “reach out” phrase (and its inflected forms) indicating someone contacting or extending communication.

    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
     ähnlichen
    -1.71
    吱
    -1.61
     ol
    -1.57
     zahlreichen
    -1.56
    顷
    -1.54
     gewährle
    -1.52
     ido
    -1.48
    ített
    -1.48
     levy
    -1.48
    ⬜
    -1.47
    POSITIVE LOGITS
    “
    2.34
    桫
    1.97
    敺
    1.96
     dirit
    1.88
     who
    1.83
    ments
    1.70
    
    1.70
    女
    1.69
    栻
    1.61
    嗞
    1.60
    Activations Density 0.002%

    No Known Activations