© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 75578
    Prev
    Next
    INDEX
    Explanations

    use of software or code

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

    The neuron fires on words that introduce or label personal evaluations or summary metrics—terms like “time,” “overall impression,” “access,” “my experience,” “satisfaction,” “use,” and “problem” that head a user’s evaluative or review statements.

    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
     recev
    -1.73
     periode
    -1.69
     aussit
    -1.67
     parut
    -1.62
     victoires
    -1.59
     rempliss
    -1.58
     brune
    -1.57
    鲧
    -1.54
    ätzen
    -1.53
     salari
    -1.53
    POSITIVE LOGITS
    :
    1.59
    ↵↵↵↵
    1.49
     in
    1.49
     after
    1.45
    _
    1.45
    视着
    1.38
     should
    1.34
    e
    1.33
    3
    1.31
    to
    1.30
    Activations Density 0.083%

    No Known Activations