© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 5806
    Prev
    Next
    INDEX
    Explanations

    at that time

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

    The neuron fires on the subordinating conjunction “that,” i.e. when introducing relative or content clauses.

    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
    χος
    0.56
     NavigationView
    0.55
     خواہ
    0.51
    িতেছে
    0.51
    asis
    0.50
    দ্ম
    0.50
     recientes
    0.50
    𝓇
    0.49
     sulle
    0.48
    ාර
    0.48
    POSITIVE LOGITS
    当时的
    3.31
    當時
    3.27
    当时
    3.22
     당시
    3.08
    那时候
    3.08
     damals
    3.06
    当時
    3.00
    那时
    2.89
    当時の
    2.83
     त्यावेळी
    2.56
    Activations Density 0.519%

    No Known Activations