© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 919
    Prev
    Next
    INDEX
    Explanations

    recommendations or advice

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

    The neuron is keyed to recognize numeric tokens expressed as floating‐point values (decimal numbers).

    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_24_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
    PhD
    0.53
    粑
    0.53
    seven
    0.51
    peptide
    0.51
    Defendants
    0.50
    7
    0.50
    Khokhlov
    0.50
    ladı
    0.50
    各项
    0.49
    ї
    0.49
    POSITIVE LOGITS
     éviter
    0.66
     evitare
    0.65
     inny
    0.63
     evitar
    0.62
     jiné
    0.61
     altijd
    0.60
     utiliser
    0.59
     refrained
    0.58
     использование
    0.58
     verwenden
    0.57
    Activations Density 0.000%

    No Known Activations