© 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. 97920
    Prev
    Next
    INDEX
    Explanations

    flow, volume, dimensional

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

    The neuron responds to specialized technical or scientific terminology—multisyllabic jargon, acronyms, hyphenated terms, and domain-specific words.

    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
    liwości
    -0.79
     fuerza
    -0.76
    NaCl
    -0.75
     Picker
    -0.73
     думал
    -0.72
     Transporter
    -0.71
    ãy
    -0.71
    彡
    -0.71
    dou
    -0.70
     AGU
    -0.69
    POSITIVE LOGITS
     hardening
    0.85
     plastic
    0.84
     evolution
    0.82
     Isles
    0.79
     Plastic
    0.78
     Evolution
    0.77
     hardened
    0.77
    ⚄
    0.75
     Huber
    0.75
     likeness
    0.75
    Activations Density 0.005%

    No Known Activations