© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 68157
    Prev
    Next
    INDEX
    Explanations

    attention-deficit and hyperactivity disorder

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

    company and product names, especially technology brands like Toshiba, and medical/psychological terms like "hyperactivity".

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

    The neuron is a specialized detector for the company name “Toshiba,” firing on its various spellings and token 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
      
    -1.64
    各样的
    -1.58
     verantwortlich
    -1.57
     upbeat
    -1.55
     acest
    -1.50
     unveiling
    -1.50
     Schuhe
    -1.49
    béco
    -1.48
    也都
    -1.45
    ⦑
    -1.45
    POSITIVE LOGITS
    乄
    2.27
    d
    1.81
    赑
    1.77
    !】
    1.73
    }$
    1.65
    涠
    1.63
    就想
    1.61
    合物
    1.55
     ткань
    1.55
    pped
    1.54
    Activations Density 0.024%

    No Known Activations