© 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. 85390
    Prev
    Next
    INDEX
    Explanations

    as possible

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

    The neuron detects comparative “as … as” intensifier constructions (e.g. “as quickly as possible,” “as accurate as possible,” etc.).

    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
     間取り
    -1.80
    彣
    -1.73
     БО
    -1.71
    Pozdrawiam
    -1.68
    ཋ
    -1.66
    ⸒
    -1.65
     rajut
    -1.60
    
    -1.55
     anúncio
    -1.52
    杙
    -1.52
    POSITIVE LOGITS
    )
    1.94
     possible
    1.66
    <bos>
    1.64
     is
    1.60
     ]
    1.59
     *
    1.57
     finest
    1.52
    ?
    1.50
    回る
    1.50
     [
    1.50
    Activations Density 0.013%

    No Known Activations