© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14986
    Prev
    Next
    INDEX
    Explanations

    plaintiffs, cytos, story, Girls, State, oak, laboratory, attention, hyper, shoppers, warrant, typo

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

    The neuron fires strongly on content‐bearing nouns (i.e. key topic words or entity nouns).

    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
     even
    -1.27
     bolster
    -1.08
     Often
    -1.08
     パス
    -1.02
     not
    -1.02
     usually
    -0.99
     both
    -0.99
     R
    -0.98
     often
    -0.97
     N
    -0.97
    POSITIVE LOGITS
    !!!!!!!
    1.13
    
    1.10
    マジで
    1.09
    相変わらず
    1.09
    filtr
    1.09
    \\
    1.08
     ungkap
    1.08
    却被
    1.06
     divertir
    1.06
     GRAND
    1.06
    Activations Density 0.065%

    No Known Activations