© 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. 111097
    Prev
    Next
    INDEX
    Explanations

    from a source

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

    The neuron strongly activates on the preposition “from,” i.e. it detects occurrences of the word “from.”

    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
     prioriti
    -1.48
    ’,
    -1.47
     caminh
    -1.37
    ’
    -1.37
    <bos>
    -1.31
     surgiu
    -1.31
     consegui
    -1.29
    .
    -1.28
     tilbud
    -1.27
    紧紧
    -1.26
    POSITIVE LOGITS
     nawet
    2.14
     by
    2.03
     him
    1.84
     даже
    1.77
     you
    1.73
    ]$.
    1.72
     admiring
    1.64
    可不是
    1.63
     обычно
    1.59
     afectadas
    1.59
    Activations Density 0.015%

    No Known Activations