© 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. 7581
    Prev
    Next
    INDEX
    Explanations

    proper nouns, place names, surnames

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

    The neuron detects place names and other geographic location references (e.g., towns, counties, roads).

    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
     again
    -1.59
    apatalk
    -1.55
    IO
    -1.48
    ius
    -1.41
    estacks
    -1.39
    2
    -1.38
                                
    -1.38
    ↵↵↵
    -1.37
    </i>
    -1.36
    нула
    -1.35
    POSITIVE LOGITS
    uskas
    1.47
    前から
    1.45
     dotted
    1.41
    acjach
    1.40
    <bos>
    1.38
    风的
    1.37
    徳島
    1.36
    Buna
    1.34
    说得
    1.34
    则是
    1.34
    Activations Density 0.431%

    No Known Activations