© 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. 590
    Prev
    Next
    INDEX
    Explanations

    computing, error, cellular, enrichment

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

    error-related terms and technical issues in code and software.

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

    This neuron spotlights code‐style identifiers and resource names (e.g. Android XML IDs, JSON keys/filenames) rather than natural language 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_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
    ,
    -1.13
     разнообраз
    -1.10
     jeśli
    -1.06
    があり
    -1.05
    的眼神
    -1.02
    ţional
    -1.02
    portug
    -1.02
     विभिन्न
    -1.01
    ēd
    -1.01
     dacă
    -1.00
    POSITIVE LOGITS
    ⅱ
    1.20
     arriva
    1.16
     інші
    1.13
    //{
    
    1.13
    Länge
    1.11
    ⽼
    1.09
    囿
    1.09
    kauft
    1.09
     ganska
    1.08
    ⅰ
    1.08
    Activations Density 0.000%

    No Known Activations