© 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. 4100
    Prev
    Next
    INDEX
    Explanations

    words beginning with letters such as bo, co, li, ex, lo, ur, phys, oct

    oai_token-act-pair · gemini-2.0-flash

    website/code snippets

    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
     They
    -1.03
     This
    -1.00
     There
    -0.95
     These
    -0.94
     The
    -0.93
    +#+#
    -0.92
    ]--;
    -0.85
     We
    -0.84
     It
    -0.81
     Finally
    -0.77
    POSITIVE LOGITS
     myself
    0.43
     iri
    0.40
     rass
    0.38
    URIComponent
    0.37
     RZ
    0.35
    izr
    0.35
     bere
    0.35
     baud
    0.35
     Jir
    0.34
    nhi
    0.34
    Activations Density 2.053%

    No Known Activations