© 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. 80714
    Prev
    Next
    INDEX
    Explanations

    comparing input to criteria

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

    The neuron flags uncommon, domain‐specific “jargon” tokens (e.g. programming identifiers like validator, pattern, message or chemistry fragments like gly, atization, bis).

    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
    预期
    -0.79
     Conroy
    -0.78
     macrophage
    -0.76
     MainAxisSize
    -0.76
     entstanden
    -0.73
    preorder
    -0.72
     gehabt
    -0.71
    AllowAnonymous
    -0.70
     mỗi
    -0.70
    criptor
    -0.68
    POSITIVE LOGITS
    Compare
    0.84
    IsValid
    0.80
     compare
    0.79
     Compare
    0.78
     Tatar
    0.77
    compare
    0.71
     Cushing
    0.70
     musste
    0.68
     comparison
    0.68
    ticism
    0.67
    Activations Density 0.021%

    No Known Activations