© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27665
    Prev
    Next
    INDEX
    Explanations

    actions followed by prepositions

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

    The neuron is picking out verbs (including past participles) and gerunds—that is, words ending in “-ed” or “-ing” that denote actions or processes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     poète
    -1.01
    ช่วย
    -0.99
    Literat
    -0.90
     speedily
    -0.88
     flesta
    -0.87
     bewegt
    -0.85
    Caracteristici
    -0.85
     微
    -0.85
    nerability
    -0.84
     giebt
    -0.84
    POSITIVE LOGITS
    grip
    1.04
     tarafından
    1.04
    zio
    1.01
     from
    1.00
     diarias
    0.98
    lijker
    0.94
    aced
    0.93
    cego
    0.91
     vanuit
    0.91
    ía
    0.90
    Activations Density 0.135%

    No Known Activations