© 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. 10-GEMMASCOPE-RES-131K
    7. 86380
    Prev
    Next
    INDEX
    Explanations

    amalgam, amalgamation, or Amal

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

    The word "amalgam" and its variations, particularly in scientific and technical contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron detects the root “amalgam” (and its variants such as “amalgamation” or “amalgamated”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     alcune
    -2.95
    蕷
    -2.78
     alcuni
    -2.69
    笵
    -2.63
     are
    -2.48
    瘊
    -2.41
    饜
    -2.39
     exceptional
    -2.38
     pulizia
    -2.38
     magnificent
    -2.36
    POSITIVE LOGITS
    During
    2.58
     burgeoning
    2.42
    When
    2.42
    Usually
    2.41
    Being
    2.41
    !!!!!
    2.39
     Making
    2.39
    Following
    2.38
     Surprisingly
    2.36
    Considering
    2.34
    Activations Density 0.002%

    No Known Activations