© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 4-GEMMASCOPE-TRANSCODER-16K
    7. 1747
    Prev
    Next
    INDEX
    Explanations

    technical descriptions of components, often using the word "attached"

    oai_token-act-pair · gemini-2.0-flash

    attached

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
     Attached
    -1.15
     attached
    -1.02
     attach
    -1.00
    attached
    -0.92
    Attached
    -0.90
     Attach
    -0.83
     Attachment
    -0.80
     ATTACH
    -0.79
     attaching
    -0.79
     attachment
    -0.79
    POSITIVE LOGITS
    tvguidetime
    1.09
    execSQL
    0.66
    ItemBackground
    0.63
     doInBackground
    0.59
    Vidite
    0.58
     puissiez
    0.58
    ümüş
    0.58
    akyti
    0.57
    ivotopis
    0.56
    AsUp
    0.56
    Activations Density 10.813%

    No Known Activations