© 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. 10-GEMMASCOPE-RES-131K
    7. 8015
    Prev
    Next
    INDEX
    Explanations

    factual findings, errors, NSDate

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

    The neuron strongly fires on multi‐syllable adjectives ending in “-al” (e.g. factual, aspectual, veridical, counterfactuals).

    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
    
    -2.83
    铪
    -2.58
    ;
    -2.47
    VIATIONS
    -2.45
     Livro
    -2.41
     Onze
    -2.33
    嘜
    -2.33
    rschein
    -2.33
     ส์
    -2.28
    🚙
    -2.28
    POSITIVE LOGITS
    7
    3.38
    4
    3.31
    6
    3.28
    3
    3.27
    re
    3.27
    1
    3.22
    8
    3.22
    _
    3.19
    2
    3.09
    Y
    2.94
    Activations Density 0.001%

    No Known Activations