© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 57122
    Prev
    Next
    INDEX
    Explanations

    financial commitments

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

    The neuron activates on numeric tokens representing explicit quantitative data (e.g. multi-digit numbers, monetary amounts, percentages).

    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
    鹑
    -0.84
    immer
    -0.79
     Dend
    -0.79
    BankAccount
    -0.79
     Fuss
    -0.72
     Nome
    -0.72
    화
    -0.70
    indexer
    -0.69
     опла
    -0.67
    atasi
    -0.67
    POSITIVE LOGITS
     commitments
    2.58
     commitment
    2.23
     committed
    2.23
     commit
    2.14
    commit
    2.08
     Commit
    2.00
     committing
    1.94
    committed
    1.94
     commits
    1.93
     Commitment
    1.92
    Activations Density 0.041%

    No Known Activations