© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113906
    Prev
    Next
    INDEX
    Explanations

    working with

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

    The neuron activates on phrases indicating collaboration or partnership, especially forms of “working with.”

    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
    いえば
    -1.42
     usadas
    -1.38
     animados
    -1.38
     häufig
    -1.37
     після
    -1.36
    といって
    -1.35
     erwäh
    -1.33
    憎
    -1.32
     mají
    -1.31
    ifying
    -1.30
    POSITIVE LOGITS
     other
    1.64
     and
    1.63
     would
    1.55
     Tüm
    1.50
     will
    1.43
    autres
    1.42
     anderen
    1.39
     nahi
    1.38
     saman
    1.37
     others
    1.35
    Activations Density 0.021%

    No Known Activations