© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 11082
    Prev
    Next
    INDEX
    Explanations

    cannot / fails

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

    The neuron fires most strongly on tokens that report failures or inability (error messages and personal “can’t/failed/unable” problem statements).

    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
     hopefully
    0.77
     fortunate
    0.77
    Hopefully
    0.74
     надеюсь
    0.73
     semoga
    0.72
     Hopefully
    0.69
    hopefully
    0.68
     অবশ্যই
    0.66
     यूं
    0.65
     عادة
    0.65
    POSITIVE LOGITS
    无法
    2.14
    無法
    2.09
     fails
    2.05
     unable
    1.95
    unable
    1.95
     failed
    1.91
    cannot
    1.90
     Unable
    1.89
     fail
    1.84
     cannot
    1.84
    Activations Density 0.334%

    No Known Activations