© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 61312
    Prev
    Next
    INDEX
    Explanations

    `from` followed by `'`

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

    The neuron specifically fires on the “from” keyword (and its surrounding import‐path context) in module import 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
    zembro
    -1.56
     conduce
    -1.45
    Additional
    -1.40
    킁
    -1.39
     eenige
    -1.34
     kaffe
    -1.34
     churrasco
    -1.34
    맑
    -1.30
     کرده
    -1.30
     новых
    -1.29
    POSITIVE LOGITS
    felpa
    1.45
     kabát
    1.45
    indonesia
    1.38
     juſ
    1.38
     många
    1.35
     However
    1.34
     nyel
    1.33
    lında
    1.30
    спомина
    1.30
     estadounidenses
    1.30
    Activations Density 0.001%

    No Known Activations