© 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. 86154
    Prev
    Next
    INDEX
    Explanations

    offspring

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

    references to offspring or direct descendants in biological or academic contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates for tokens that are complete standalone words (i.e. not subword fragments).

    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
    .
    -3.27
    mathrm
    -2.88
    }
    -2.84
    蛲
    -2.77
    $
    -2.72
     '
    -2.69
    gefügt
    -2.66
    嵛
    -2.64
    
    -2.64
     anunciar
    -2.58
    POSITIVE LOGITS
    
    3.19
    灩
    2.50
    e
    2.48
    􀂃
    2.48
    replyreport
    2.44
    🕗
    2.42
    งิน
    2.42
    馀
    2.38
    いますが
    2.36
    霂
    2.25
    Activations Density 0.003%

    No Known Activations