© 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-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 9831
    Prev
    Next
    INDEX
    Explanations

    a variety of function words and pronouns, often appearing close to verbs, with a focus on the wanting/hearing/being told.

    oai_token-act-pair · gemini-2.0-flash

    hearing/listening

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     ProtoMessage
    -0.69
    ⓧ
    -0.65
     protoimpl
    -0.63
     BoxFit
    -0.59
     Mach
    -0.58
    MENAFN
    -0.57
    getColumnIndex
    -0.54
    afficheront
    -0.54
    }{*}{
    -0.52
    NOPQRST
    -0.52
    POSITIVE LOGITS
     haer
    0.54
    äsident
    0.53
     rédaction
    0.52
     picioare
    0.52
    hassee
    0.52
     morire
    0.52
     bénéfices
    0.50
     ligiloj
    0.50
     facie
    0.48
    ThroughAttribute
    0.48
    Activations Density 1.743%

    No Known Activations