© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 101084
    Prev
    Next
    INDEX
    Explanations

    names-for-boys.com/names/boys/B/Be/Ben...sic-Rock-Songs"- "Johns Hopkins University - Essays that Worked... After that experience, my daughter put together her essays easily with my son’"Let's summarize the findings:- `TOP_POSITIVE_LOGITS`: Contains medical terms (vermectin,cherichia), foreign languages (Chinese, Japanese, German). This suggests the neuron might be related to specific technical terms or international contexts. names.com gender neutral

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

    The neuron activates on numeric tokens—especially those containing digits with decimal points.

    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
     and
    -2.08
    Our
    -1.87
    我們的
    -1.59
    If
    -1.56
    Boletín
    -1.50
     our
    -1.50
    低声
    -1.48
     nosso
    -1.41
    We
    -1.38
    縝
    -1.30
    POSITIVE LOGITS
     needed
    1.38
    cherichia
    1.34
    ܜ
    1.30
    vermectin
    1.30
    是什么意思
    1.29
    もあった
    1.29
    这种
    1.23
    delmi
    1.23
    也没
    1.23
     ermöglicht
    1.23
    Activations Density 0.013%

    No Known Activations