© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1829
    Prev
    Next
    INDEX
    Explanations

    rotation importance

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

    The neuron fires strongly on specialized technical or engineering–style nouns and jargon terms.

    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
    WHEREAS
    0.60
     Luật
    0.58
    ฬ
    0.55
     세
    0.55
    인데요
    0.55
    മാണ്
    0.55
     Sở
    0.54
     サ
    0.54
     Whereas
    0.53
     。
    0.53
    POSITIVE LOGITS
     весьма
    1.46
     بسیار
    1.38
     খুবই
    1.37
     oldukça
    1.36
     extremely
    1.30
    extremely
    1.27
     बेहद
    1.26
     able
    1.25
     매우
    1.25
     상당히
    1.21
    Activations Density 0.712%

    No Known Activations