© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14125
    Prev
    Next
    INDEX
    Explanations

    requirements (

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

    The neuron is tuned to pick out key content words—especially verbs (often in past‐participle or simple‐past form) and salient adverbs or temporal connectors—rather than common function words.

    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
     conspicuously
    0.67
    jord
    0.64
    ++)//
    0.64
    পর্য
    0.61
    আরো
    0.61
     dépasse
    0.59
    䨘
    0.59
    );
    0.58
     emphatically
    0.58
    আবার
    0.57
    POSITIVE LOGITS
    н
    1.06
    l
    0.98
    n
    0.88
    nd
    0.82
     Parr
    0.82
    м
    0.82
    ю
    0.80
    jų
    0.79
     Vì
    0.78
    lendir
    0.78
    Activations Density 0.000%

    No Known Activations