© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 54327
    Prev
    Next
    INDEX
    Explanations

    fine-grained control

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

    The neuron activates on words or characters denoting “fine” or fine-grained detail (e.g. “fine,” “finer,” “grained,” Chinese “细”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     немного
    -0.69
     Summar
    -0.69
    ITATION
    -0.69
    asymp
    -0.68
     Rugg
    -0.68
    արտ
    -0.66
    SafeAreaView
    -0.66
    izzi
    -0.65
     ساده
    -0.65
     basics
    -0.65
    POSITIVE LOGITS
     finer
    4.28
     fine
    4.16
    fine
    3.55
    Fine
    3.42
    细
    3.16
    細
    3.09
     finely
    3.08
     Fine
    3.08
    FINE
    2.98
     detailed
    2.95
    Activations Density 0.170%

    No Known Activations