© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39887
    Prev
    Next
    INDEX
    Explanations

    teddy bear, pretrained model, Supplementary figures

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

    The neuron fires on press‐release metadata and section headers (e.g. wire service tags like “PRNewswire,” “CNW,” and labels such as “Supplementary”).

    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
    1
    -3.06
    蛧
    -2.45
    同じ
    -2.30
    ?"
    -2.28
     standout
    -2.27
    a
    -2.25
    甚至
    -2.17
    籥
    -2.16
    Ꝑ
    -2.16
     cardinals
    -2.13
    POSITIVE LOGITS
     técn
    2.20
    鄠
    2.13
    </td>
    2.13
    是什么意思
    2.11
     Jogo
    1.99
     Beautiful
    1.91
     systeem
    1.89
     Our
    1.88
     奶茶
    1.88
     lavander
    1.88
    Activations Density 0.001%

    No Known Activations