© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66937
    Prev
    Next
    INDEX
    Explanations

    events and outcomes

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

    The neuron fires most strongly on verbs and nouns that describe the key outcomes or dramatic effects in a narrative (e.g. “nothing happened,” “damage,” “ground to a halt,” “unusable”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     your
    -2.03
     their
    -1.68
     you
    -1.63
     budete
    -1.53
     tráiler
    -1.46
     menyukai
    -1.38
    bakter
    -1.38
    واف
    -1.34
    Zusammen
    -1.33
     السياس
    -1.32
    POSITIVE LOGITS
     was
    2.66
     didn
    1.94
     وكان
    1.53
     était
    1.47
     وكانت
    1.36
     étaient
    1.35
     estava
    1.32
     I
    1.31
    дава
    1.28
     threw
    1.27
    Activations Density 0.036%

    No Known Activations