© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 69200
    Prev
    Next
    INDEX
    Explanations

    lack of available resources

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find words and phrases expressing quantity or amount (e.g., few, many, any, just, really, available).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     aldrig
    -0.93
    itinéraire
    -0.90
    newtheorem
    -0.86
    interp
    -0.83
     asos
    -0.82
    يرا
    -0.82
     durchaus
    -0.82
     никогда
    -0.81
    audio
    -0.81
    skine
    -0.81
    POSITIVE LOGITS
     options
    1.74
     available
    1.73
     suitable
    1.59
     alternatives
    1.34
     for
    1.27
     viable
    1.18
     adequate
    1.13
     подходя
    1.13
     availability
    1.10
     readily
    1.05
    Activations Density 0.088%

    No Known Activations