© 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. 34-GEMMASCOPE-RES-131K
    4. 84281
    Prev
    Next
    INDEX
    Explanations

    Present study

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

    The neuron fires on specialized technical or scientific tokens—i.e. jargon and identifiers such as study‐reporting words (“present”), acronyms with numbers or hyphens (CTLA-4, PD-1), code identifiers or file extensions (Dtype, .hpp), and similar domain‐specific terms.

    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
    bliž
    -0.74
     EVERY
    -0.73
    SceneManager
    -0.70
     прожи
    -0.69
     attempting
    -0.69
     Respiratory
    -0.68
     membeli
    -0.67
    ambahkan
    -0.66
     Neue
    -0.66
    issenschaft
    -0.65
    POSITIVE LOGITS
    sheets
    0.73
     inten
    0.72
    elds
    0.72
    ceptor
    0.71
     attr
    0.71
    Premium
    0.68
     elves
    0.68
    iffen
    0.68
     cheeks
    0.68
     rush
    0.67
    Activations Density 0.050%

    No Known Activations