© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41235
    Prev
    Next
    INDEX
    Explanations

    avoiding negative outcomes

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

    The neuron strongly activates on occurrences of the word “avoid.”

    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
     to
    -1.67
     vermutlich
    -1.59
     on
    -1.48
     wahrscheinlich
    -1.40
     demnach
    -1.38
    男朋友
    -1.34
    ęcie
    -1.32
    )
    -1.30
    ajuku
    -1.30
     ferner
    -1.29
    POSITIVE LOGITS
    鼇
    1.63
     pinggang
    1.62
    Atlético
    1.57
    
    1.51
    ʏ
    1.51
    
    1.50
    ଃ
    1.48
     グリル
    1.46
    Siempre
    1.45
    niños
    1.43
    Activations Density 0.024%

    No Known Activations