© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 6885
    Prev
    Next
    INDEX
    Explanations

    words related to creating something, especially sewing or crafts.

    oai_token-act-pair · gemini-2.0-flash

    something

    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
     Someone
    -1.13
    Someone
    -1.12
     someone
    -1.00
    someone
    -0.89
     quelqu
    -0.68
     SOMEONE
    -0.66
    Jereo
    -0.65
     iemand
    -0.61
    دانشنامهٔ
    -0.60
     Somebody
    -0.60
    POSITIVE LOGITS
     something
    2.28
    something
    2.03
    Something
    1.79
     Something
    1.77
     algo
    1.38
     SOMETHING
    1.32
     quelque
    1.30
     något
    1.25
     qualcosa
    1.23
     iets
    1.22
    Activations Density 3.188%

    No Known Activations