© 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. 31-GEMMASCOPE-2-RES-16K
    7. 216
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on direct second‐person address and associated action verbs—i.e. “you/your” plus verbs like helps, gives, share, reflect, etc., marking reader-directed guidance.

    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
     Use
    0.42
     =
    0.41
     $=\
    0.41
     $
    0.41
     $\
    0.41
     $=
    0.41
     Committee
    0.40
     Section
    0.38
     Sabha
    0.38
     $=$
    0.38
    POSITIVE LOGITS
     fantastical
    0.91
     quirky
    0.85
     melodies
    0.84
     soulful
    0.80
     allerlei
    0.79
     histórias
    0.78
     masterpieces
    0.76
     evocative
    0.76
     wildflowers
    0.75
     whimsical
    0.75
    Activations Density 9.473%

    No Known Activations