© 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. 15079
    Prev
    Next
    INDEX
    Explanations

    the words "to" "at", "go", and "see".

    oai_token-act-pair · gemini-2.0-flash

    at/to

    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
     attend
    -0.55
    shop
    -0.53
     attending
    -0.52
    club
    -0.50
    school
    -0.49
    place
    -0.48
    church
    -0.48
    netje
    -0.48
     office
    -0.47
     museum
    -0.47
    POSITIVE LOGITS
    providedIn
    0.75
    Datuak
    0.71
    GEBURTSDATUM
    0.68
    Geplaatst
    0.68
    ValueStyle
    0.68
     onPostExecute
    0.64
    abestanden
    0.63
     TextInputType
    0.62
     Signalez
    0.58
     AppModule
    0.58
    Activations Density 4.786%

    No Known Activations