© 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. 319
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on first- and second-person pronouns (e.g. “I,” “you,” “we,” “ourselves”).

    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
     youll
    0.30
    ).
    0.28
     arti
    0.27
    0
    0.27
     flute
    0.26
     olursa
    0.26
    }+{
    0.26
     towards
    0.26
    }.
    0.26
     spp
    0.25
    POSITIVE LOGITS
     need
    0.61
     NEED
    0.55
     mustn
    0.54
     ought
    0.51
    need
    0.50
     owe
    0.49
     shouldn
    0.49
     needn
    0.48
    必须
    0.47
     necesidad
    0.47
    Activations Density 0.832%

    No Known Activations