© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 54232
    Prev
    Next
    INDEX
    Explanations

    mathematical inequalities (`<`, `<=`)

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

    The neuron strongly activates on sequences of uppercase letters (all‐caps words or acronyms).

    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
     of
    -1.22
    exercice
    -1.17
     всех
    -1.03
     from
    -1.00
    ſſen
    -0.97
    都是
    -0.97
     efisien
    -0.94
    Therefore
    -0.94
     lett
    -0.93
    alers
    -0.93
    POSITIVE LOGITS
     or
    1.38
     you
    1.06
     predominant
    1.03
     zamiast
    1.02
    Мини
    1.02
    illère
    1.01
    窣
    0.98
     your
    0.98
     him
    0.98
    vestream
    0.96
    Activations Density 0.001%

    No Known Activations