© 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. 73277
    Prev
    Next
    INDEX
    Explanations

    individual things

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

    This neuron detects formal, information‐dense reporting language—especially passages packed with statistics, measurements, and other quantitative details.

    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
     เขา
    -0.90
     которые
    -0.89
    ofollow
    -0.85
    andidate
    -0.84
    parque
    -0.83
     supuestamente
    -0.79
    pagnole
    -0.79
     остров
    -0.78
    vacy
    -0.77
     responses
    -0.77
    POSITIVE LOGITS
     its
    1.73
     individual
    1.45
    Individual
    1.27
    individual
    1.23
     individually
    1.09
     itself
    1.06
     individuales
    1.05
    单独
    1.05
    Whole
    1.05
     separate
    1.05
    Activations Density 0.172%

    No Known Activations