© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 395
    Prev
    Next
    INDEX
    Explanations

    going back again

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

    This neuron fires on numeric tokens—i.e. pieces of text that represent numbers or numeric constants.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    最后的
    -0.95
     exámenes
    -0.87
    ֭
    -0.82
     Entscheidungen
    -0.81
    mif
    -0.81
     bestehende
    -0.79
     répond
    -0.76
     Confirm
    -0.76
    それぞれの
    -0.75
    respond
    -0.75
    POSITIVE LOGITS
     again
    1.59
     снова
    1.25
    again
    1.20
    Again
    1.16
     another
    1.13
     wieder
    1.07
    に戻
    1.06
     다시
    1.05
     دوباره
    1.04
    重新
    1.02
    Activations Density 0.009%

    No Known Activations