© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 76556
    Prev
    Next
    INDEX
    Explanations

    actions involving him or herself

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

    The neuron flags words expressing internal states or attitudes—especially mental‐state verbs (e.g. wanted, considered, believing).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    娶
    -1.53
     boyhood
    -1.34
     istrinya
    -1.30
    妻
    -1.30
    给她
    -1.23
     its
    -1.20
     établie
    -1.18
    其
    -1.13
    对她
    -1.13
     himself
    -1.09
    POSITIVE LOGITS
     herself
    5.16
    herself
    3.58
     сама
    2.25
     نفسها
    2.00
     powinna
    1.55
     elle
    1.51
     została
    1.40
     должна
    1.40
     him
    1.38
    сама
    1.37
    Activations Density 0.045%

    No Known Activations