© 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. 10-GEMMASCOPE-RES-131K
    7. 105431
    Prev
    Next
    INDEX
    Explanations

    followed by descriptive words

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

    The neuron is selectively detecting words that indicate spatial location or orientation (e.g., inside, down, below, above, at).

    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
     will
    -2.23
     more
    -2.00
     this
    -1.97
     any
    -1.92
     then
    -1.89
     as
    -1.64
     most
    -1.59
     in
    -1.57
     it
    -1.52
     even
    -1.49
    POSITIVE LOGITS
    mektedir
    1.48
     них
    1.41
     domine
    1.41
     ALREADY
    1.36
     itself
    1.36
     wollt
    1.35
    ='';
    1.34
     appelé
    1.34
    えています
    1.34
    ниц
    1.31
    Activations Density 0.054%

    No Known Activations