© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 18-GEMMASCOPE-TRANSCODER-16K
    7. 8072
    Prev
    Next
    INDEX
    Explanations

    clauses beginning with "which is why" or "for this reason"

    oai_token-act-pair · gemini-2.0-flash

    That's/Therefore

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_18/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.18.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
    bcryptjs
    -0.46
    engers
    -0.44
     so
    -0.43
    jst
    -0.41
    mybatisplus
    -0.41
    olen
    -0.41
    Collegamenti
    -0.40
    IVENESS
    -0.39
    mpion
    -0.39
    بش
    -0.39
    POSITIVE LOGITS
     thats
    1.10
     Daarom
    1.09
     Thats
    1.08
     Derfor
    1.02
     därför
    1.00
     Dlatego
    0.99
    Thats
    0.98
    latego
    0.97
    thats
    0.97
     dlatego
    0.95
    Activations Density 0.778%

    No Known Activations