© 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. 772
    Prev
    Next
    INDEX
    Explanations

    the various

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

    The neuron is spotting mid-frequency content words—that is, domain-specific or otherwise informational terms with high TF-IDF importance.

    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
     bijvoorbeeld
    -1.20
     esimerkiksi
    -1.20
    erweise
    -1.14
     например
    -1.12
     could
    -1.09
     any
    -1.08
     will
    -1.05
     or
    -1.03
     an
    -1.01
     if
    -0.98
    POSITIVE LOGITS
     wichtigsten
    1.92
     various
    1.76
     différents
    1.70
     highlights
    1.69
     principaux
    1.68
     salient
    1.65
    various
    1.63
    highlights
    1.61
     основные
    1.61
     principales
    1.61
    Activations Density 0.169%

    No Known Activations