© 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. 22-GEMMASCOPE-RES-131K
    7. 54375
    Prev
    Next
    INDEX
    Explanations

    loading data configurations

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

    The neuron fires on object or map key names in code—i.e. identifiers immediately preceding a colon in JSON‐ or JavaScript‐style key–value pairs.

    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
    ől
    -1.08
    ()!=
    -1.04
    MediaStore
    -1.02
    。
    -1.01
    stadion
    -1.00
    γραφία
    -1.00
    }>
    -1.00
    部に
    -0.99
    可能
    -0.99
     frå
    -0.98
    POSITIVE LOGITS
    がありますが
    1.24
    時間も
    1.13
    爆笑
    1.05
    そこは
    1.05
     vanguardia
    1.04
    了一個
    1.02
    曲は
    1.02
     адресу
    1.00
    Ecco
    0.99
     vaše
    0.98
    Activations Density 0.018%

    No Known Activations