© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11468
    Prev
    Next
    INDEX
    Explanations

    the author

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

    The neuron selectively activates on multi-letter uppercase abbreviations or acronyms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    LR
    0.50
    JSX
    0.46
    Sono
    0.45
    (
    0.44
     (
    0.44
    BP
    0.43
    }(\
    0.43
    ฯ
    0.43
    arem
    0.42
    orsch
    0.42
    POSITIVE LOGITS
     july
    0.49
     prosent
    0.47
     gallant
    0.46
     seclusion
    0.46
     singkat
    0.45
     McCann
    0.44
     waistcoat
    0.43
     hv
    0.42
     euh
    0.42
    rifft
    0.42
    Activations Density 0.127%

    No Known Activations