© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 7062
    Prev
    Next
    INDEX
    Explanations

    numbers with a hyphen after them

    oai_token-act-pair · gemini-2.0-flash

    -

    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
     ſeveral
    -0.91
     Theſe
    -0.85
     Efq
    -0.85
     Reſ
    -0.85
    ſelf
    -0.84
     snippetHide
    -0.83
     étoient
    -0.83
     fevere
    -0.83
     themſelves
    -0.82
     avoient
    -0.79
    POSITIVE LOGITS
     week
    0.52
     day
    0.49
    stdc
    0.44
     Par
    0.43
     WEEK
    0.42
     Mont
    0.41
     
    0.41
     Day
    0.41
     par
    0.40
     night
    0.40
    Activations Density 0.492%

    No Known Activations