© 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. 34-GEMMASCOPE-RES-131K
    7. 82449
    Prev
    Next
    INDEX
    Explanations

    wavelength range

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

    The neuron is effectively dead—it never activates on any tokens in these excerpts and thus doesn’t detect any pattern.

    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
     always
    -1.13
     knew
    -1.06
     from
    -1.06
     mostly
    -1.03
     certainly
    -1.01
     only
    -1.01
     other
    -1.00
     definitely
    -0.97
     unequivocally
    -0.96
    امن
    -0.95
    POSITIVE LOGITS
     MAX
    0.98
     رؤية
    0.90
    geführt
    0.90
    neden
    0.90
    esss
    0.90
    IDK
    0.90
     hinauf
    0.90
    嚀
    0.89
     max
    0.88
     …..
    0.88
    Activations Density 0.085%

    No Known Activations