© 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. 562
    Prev
    Next
    INDEX
    Explanations

    a wide range of words, but appears to be somewhat biased towards political and societal commentary.

    oai_token-act-pair · gemini-2.0-flash

    Period

    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
     genoux
    -0.48
     tiroirs
    -0.36
     lèvres
    -0.35
     boî
    -0.35
     dalamnya
    -0.35
     gouttes
    -0.35
     manutenzione
    -0.35
     rayures
    -0.34
     engraçadas
    -0.34
     cœurs
    -0.33
    POSITIVE LOGITS
    ]--;
    0.35
    σθαι
    0.32
    ignty
    0.32
    [::-
    0.32
    sterious
    0.31
    Становништво
    0.30
    ophyll
    0.30
     breeches
    0.30
    DoubleQuotes
    0.29
    thmetic
    0.29
    Activations Density 16.371%

    No Known Activations