© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 760
    Prev
    Next
    INDEX
    Explanations

    The neuron primarily activates on words and phrases describing dynamic movement or flow (e.g., transition, flow, handling, breaststroke).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    phrases describing the movement, transfer, or flow of physical materials or substances.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    phrases about the movement or handling of things (data, goods, people, quantities), emphasizing collection, flow, transfer, and delivery relationships.

    oai_token-act-pair · gpt-5Triggered 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
    ospitals
    0.71
    élevage
    0.71
    comboBox
    0.70
    atedral
    0.67
     monasteries
    0.67
    enerbah
    0.64
     музея
    0.64
     отношений
    0.63
    FBSDKLogin
    0.63
    ComboBox
    0.63
    POSITIVE LOGITS
     materials
    1.64
     material
    1.46
     foodstuffs
    1.45
     goods
    1.45
     substances
    1.45
     nutrients
    1.44
     fluids
    1.43
     материал
    1.41
     вещества
    1.40
     materiál
    1.39
    Activations Density 2.286%

    No Known Activations