© 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. 22-GEMMASCOPE-RES-131K
    4. 12850
    Prev
    Next
    INDEX
    Explanations

    you may

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

    The neuron activates on second-person address and imperative instruction words (e.g. “you” and “Change”) in documentation or code comments.

    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
     its
    -1.70
     anyone
    -1.20
     their
    -1.20
     any
    -1.16
     nombreuses
    -1.13
     belga
    -1.09
    länge
    -1.09
     alleen
    -1.08
    Là
    -1.08
     način
    -1.07
    POSITIVE LOGITS
    camento
    1.28
     Поэтому
    1.12
     UITableViewCell
    1.11
    仿佛
    1.11
     поэтому
    1.09
    なので
    1.07
    pernicus
    1.07
    玫
    1.05
     Therefore
    1.02
    versiti
    1.02
    Activations Density 0.001%

    No Known Activations