© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 128362
    Prev
    Next
    INDEX
    Explanations

    peddling falsehoods or beliefs

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

    The neuron detects verbs that convey forcefully imposing, pushing, or foisting something on others.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     and
    -1.57
     centred
    -1.45
     centered
    -1.38
     finally
    -1.37
    følge
    -1.21
    . 
    -1.20
    と思われる
    -1.18
    该公司
    -1.16
     駅
    -1.16
     vinilo
    -1.13
    POSITIVE LOGITS
     menilai
    1.66
     memeriksa
    1.59
     mempel
    1.48
     Specifically
    1.44
     berkomunikasi
    1.42
     melanjutkan
    1.41
     meneg
    1.41
     mengucap
    1.39
    </h2>
    1.38
     menyambut
    1.37
    Activations Density 0.047%

    No Known Activations