© 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. 14811
    Prev
    Next
    INDEX
    Explanations

    conjunctions or disjunctions

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

    The neuron is triggered by unusually long, multi‐character tokens (i.e. lengthy words).

    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
    になれば
    -0.90
    の人は
    -0.85
    になれる
    -0.84
    logues
    -0.82
    の方法
    -0.80
     Tennessee
    -0.80
     Schreiben
    -0.80
     I
    -0.80
     осо
    -0.79
     ficar
    -0.79
    POSITIVE LOGITS
     maupun
    1.33
     and
    1.29
     மற்றும்
    1.13
    以及
    1.09
    および
    1.07
     oraz
    1.03
     or
    1.01
     ataupun
    0.98
     및
    0.94
    または
    0.94
    Activations Density 0.044%

    No Known Activations