© 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. 66588
    Prev
    Next
    INDEX
    Explanations

    found in

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

    The neuron fires on occurrences of the word “found,” as in “found in …” or “found …” in metadata or reference contexts.

    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
     Когда
    -2.86
    time
    -2.70
    ↵
    -2.59
    吭
    -2.56
    ',
    -2.53
    ⚋
    -2.53
     После
    -2.31
     当
    -2.31
     incite
    -2.30
     إنها
    -2.25
    POSITIVE LOGITS
    什么事
    3.00
    乆
    2.84
    但是在
    2.81
    縂
    2.80
    
    2.77
     fashioned
    2.75
     policiales
    2.67
    所以我
    2.67
     unangemess
    2.64
    逖
    2.63
    Activations Density 0.036%

    No Known Activations