© 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. 74552
    Prev
    Next
    INDEX
    Explanations

    iceberg, sulfur, WrestleMania, Besar, cordova

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

    The neuron strongly fires on each occurrence of “iceberg” (e.g. the idiom “tip of the iceberg,” product names, and “iceberg lettuce”), effectively detecting that specific word.

    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
    }$
    -2.56
    ⑊
    -2.47
    閦
    -2.44
    泚
    -2.36
     ſou
    -2.33
    wiście
    -2.28
    涠
    -2.25
    ⢊
    -2.23
    
    -2.22
    醅
    -2.20
    POSITIVE LOGITS
    </h2>
    2.78
     ۸
    2.36
    あまり
    2.36
     schwierig
    2.33
     pleas
    2.33
     midst
    2.19
    最重要的
    2.14
    -
    2.14
     In
    2.13
    \
    2.11
    Activations Density 0.003%

    No Known Activations