© 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. 34-GEMMASCOPE-RES-131K
    7. 64366
    Prev
    Next
    INDEX
    Explanations

    code examples and tests

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

    The neuron is detecting dot-access expressions in code—that is, occurrences of the “.” token used for method calls or property lookups.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     fordert
    -1.06
     if
    -1.06
     confirmado
    -0.96
    arraycopy
    -0.92
    ﱠ
    -0.91
     jeśli
    -0.91
    getOutputStream
    -0.91
     млн
    -0.91
     चीज़
    -0.90
     their
    -0.90
    POSITIVE LOGITS
     exerci
    1.09
    ={()
    1.07
     test
    1.02
    test
    1.02
     sut
    1.01
     simula
    0.99
    Mock
    0.99
     Test
    0.98
     teste
    0.98
    teste
    0.96
    Activations Density 0.026%

    No Known Activations