© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 3726
    Prev
    Next
    INDEX
    Explanations

    descriptions of used merchandise

    oai_token-act-pair · gemini-2.0-flash

    physical damage indicators

    np_token-act-pair-logits · gpt-4o-mini

    bumps and dents

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
    tanleria
    -0.82
     насељу
    -0.78
     disambiguazione
    -0.73
    saraba
    -0.72
    ########.
    -0.65
     kaarangay
    -0.63
    انيف
    -0.62
    abetes
    -0.62
    ContentAsync
    -0.60
     saites
    -0.60
    POSITIVE LOGITS
     knocks
    0.76
     bumped
    0.67
     den
    0.66
     ding
    0.65
     knocked
    0.63
     dent
    0.63
     dents
    0.63
    剐
    0.62
     bump
    0.62
     knock
    0.61
    Activations Density 1.362%

    No Known Activations