© 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. 99751
    Prev
    Next
    INDEX
    Explanations

    order before

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

    The neuron selectively activates on capitalized words in proper names (as in “Gods” from “Valley of the Gods”).

    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
     outros
    -1.70
     “
    -1.68
     Other
    -1.63
     digna
    -1.58
    Ꮟ
    -1.56
    ſelves
    -1.50
    誊
    -1.48
    与此
    -1.47
    のよう
    -1.47
     want
    -1.45
    POSITIVE LOGITS
    2
    2.53
    8
    2.39
    4
    2.36
    9
    2.34
    6
    2.22
    5
    2.19
    3
    2.19
    0
    1.99
    7
    1.97
    และ
    1.88
    Activations Density 0.000%

    No Known Activations