© 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
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 14425
    Prev
    Next
    INDEX
    Explanations

    This neuron seems to activate on a variety of words, with some preference towards food items, French words, and opera terminology.

    oai_token-act-pair · gemini-2.0-flash

    varied text snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     for
    -0.48
     the
    -0.48
    LookAnd
    -0.47
    SequentialGroup
    -0.47
    حياته
    -0.45
     Jurí
    -0.43
    følgelig
    -0.43
    iedene
    -0.43
    Kedua
    -0.42
    calling
    -0.42
    POSITIVE LOGITS
     ErrIntOverflow
    0.59
     CreateTagHelper
    0.56
    曖昧さ回避
    0.55
     تانيه
    0.53
     صوتيه
    0.51
    substack
    0.51
    ppr
    0.50
    czegóły
    0.50
    entery
    0.50
    rxjs
    0.50
    Activations Density 0.816%

    No Known Activations