© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91493
    Prev
    Next
    INDEX
    Explanations

    behavior of people

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

    The neuron strongly activates on content-bearing words (nouns and verbs) that carry topical or informational weight, while ignoring common function 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_34/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
    #\
    -0.94
    providedIn
    -0.94
    serv
    -0.93
     cy
    -0.92
    edere
    -0.91
     vere
    -0.91
    سن
    -0.90
    chium
    -0.90
    ichi
    -0.89
     figura
    -0.88
    POSITIVE LOGITS
     tend
    2.08
     often
    2.08
     usually
    1.96
     always
    1.87
     tends
    1.66
     prefer
    1.43
     think
    1.38
     sometimes
    1.38
     believe
    1.36
     tended
    1.34
    Activations Density 0.078%

    No Known Activations