© 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. 71058
    Prev
    Next
    INDEX
    Explanations

    locations and technical terms

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

    The neuron fires on occurrences of specific product, brand, or technical item names (e.g. textile companies, equipment models, sheet-music titles, ICSP device identifiers).

    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
    Pending
    -0.96
    ⇤
    -0.93
    AKAN
    -0.92
     befind
    -0.89
    cja
    -0.89
    itize
    -0.88
     ワ
    -0.85
    Login
    -0.84
    RTHOOK
    -0.83
     trening
    -0.83
    POSITIVE LOGITS
     setelah
    1.02
    Since
    0.99
    Without
    0.97
     without
    0.96
    Initially
    0.94
    Depending
    0.93
    During
    0.93
    Almost
    0.91
    These
    0.90
    Ironically
    0.90
    Activations Density 0.026%

    No Known Activations