© 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. 13065
    Prev
    Next
    INDEX
    Explanations

    except

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

    The neuron strongly activates on runs of question‐mark punctuation (e.g. “???”).

    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
    Utils
    0.52
     Utils
    0.49
     utils
    0.48
    /__
    0.47
    utils
    0.45
     Hooker
    0.45
     Massachusetts
    0.45
     Nag
    0.44
     |=
    0.44
     Matsumoto
    0.44
    POSITIVE LOGITS
     えっと
    0.55
    --------
    0.53
    oser
    0.53
                  
    0.52
     なく
    0.52
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.51
    שה
    0.51
    ………..
    0.51
               
    0.50
     ..........
    0.50
    Activations Density 0.000%

    No Known Activations