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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129271
    Prev
    Next
    INDEX
    Explanations

    t followed by numbers or codes

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

    This neuron responds to isolated single-letter identifiers (e.g. variable names or one-letter tags) appearing in formulas or code/XML.

    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
    Mereka
    -1.54
    s
    -1.53
     torner
    -1.45
    𖥦
    -1.45
    强调
    -1.42
     combing
    -1.40
    和她
    -1.39
    iconductors
    -1.38
    鹟
    -1.37
     تغییر
    -1.36
    POSITIVE LOGITS
     we
    2.19
     Its
    1.85
     også
    1.77
     também
    1.66
     また
    1.65
    しまった
    1.62
    籥
    1.53
    慥
    1.50
    cillas
    1.50
     jego
    1.45
    Activations Density 0.079%

    No Known Activations