© 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. 2096
    Prev
    Next
    INDEX
    Explanations

    relative clauses or translations

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

    The neuron flags segments describing methods or procedures—that is, step‐by‐step, technical or instructional passages.

    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
     variously
    1.28
     doubtless
    1.20
     ostensibly
    1.15
     nominally
    1.13
     ;-)
    1.11
     putative
    1.10
     invariably
    1.06
     nahin
    1.06
     hapless
    1.04
     erstwhile
    1.02
    POSITIVE LOGITS
     aswell
    1.34
     advancements
    1.33
    recieve
    1.32
    됬
    1.32
    또한
    1.25
     extremamente
    1.20
     देखील
    1.18
     minimalistic
    1.18
     recieve
    1.18
     recieved
    1.17
    Activations Density 1.851%

    No Known Activations