© 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. 20304
    Prev
    Next
    INDEX
    Explanations

    density of states

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

    The neuron is specialized to pick out multiword “[noun] of [noun]” phrases (e.g. “density of states,” “equation of motion,” “loss of function”).

    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
    This
    -2.11
    If
    -2.02
    How
    -1.95
    You
    -1.93
    what
    -1.91
    When
    -1.89
    Surprisingly
    -1.84
    It
    -1.81
     two
    -1.76
    Where
    -1.76
    POSITIVE LOGITS
     également
    2.00
    珝
    1.88
     ก็
    1.87
    灩
    1.87
     也
    1.84
    étés
    1.83
    いただきたい
    1.82
    OCOLATE
    1.81
     również
    1.79
    淬
    1.78
    Activations Density 0.113%

    No Known Activations