© 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. 7948
    Prev
    Next
    INDEX
    Explanations

    perhaps or could

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

    The neuron gloms onto action verbs in procedural or how-to contexts (e.g. dyed, restrict, include, export, loaded, allowed).

    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
     Luật
    0.79
     エネルギー
    0.77
     Ernährung
    0.75
     kttt
    0.75
     顔
    0.73
     Gesamt
    0.73
     sufrimiento
    0.73
     consequências
    0.72
     değildir
    0.70
     層
    0.70
    POSITIVE LOGITS
    can
    0.82
     reference
    0.79
    may
    0.78
     may
    0.77
     might
    0.75
    might
    0.74
    可以在
    0.73
     could
    0.73
    可以用
    0.73
    用に
    0.73
    Activations Density 0.187%

    No Known Activations