© 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. 34-GEMMASCOPE-RES-131K
    7. 108
    Prev
    Next
    INDEX
    Explanations

    testing assertions

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

    This neuron activates on numeric sequences in code or comments, i.e. literal numbers.

    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
     tests
    -0.89
    」,
    -0.84
    zehn
    -0.83
    ziem
    -0.82
     test
    -0.82
     their
    -0.79
    acceptable
    -0.79
    ugier
    -0.77
    fehl
    -0.76
     romántico
    -0.76
    POSITIVE LOGITS
     believe
    1.23
    assertEquals
    1.12
     Believe
    0.94
    mark
    0.91
     belief
    0.90
    Believe
    0.87
    assertNotNull
    0.86
     beliefs
    0.84
    assert
    0.81
    atterns
    0.78
    Activations Density 0.001%

    No Known Activations