© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44698
    Prev
    Next
    INDEX
    Explanations

    turned into a/an/the

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

    This neuron primarily activates on the phrase “turn (or its inflected forms) into.”

    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
    9
    -2.72
    7
    -2.06
    8
    -2.06
    6
    -2.05
    despite
    -2.05
     FRENCH
    -2.03
    
    -1.98
     ANCI
    -1.94
    2
    -1.91
    quirer
    -1.90
    POSITIVE LOGITS
     a
    2.61
     ‘
    2.09
     vivido
    2.05
    a
    1.87
     something
    1.85
     –
    1.84
     aprendido
    1.80
     “
    1.78
     leído
    1.77
     payé
    1.76
    Activations Density 0.008%

    No Known Activations