© 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. 6649
    Prev
    Next
    INDEX
    Explanations

    a combination of short gibberish words and longer words related to proximity

    oai_token-act-pair · gemini-2.0-flash

    z

    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
     approach
    -2.52
    approach
    -2.36
     Approach
    -2.36
     APPROACH
    -2.28
     approaches
    -2.27
    Approach
    -2.13
     approached
    -2.11
     Approaches
    -1.90
     approaching
    -1.87
     approche
    -1.86
    POSITIVE LOGITS
     équipée
    0.57
     auffi
    0.57
    lismo
    0.56
     élas
    0.55
     compleanno
    0.54
     ainfi
    0.54
     itſelf
    0.53
     Ahead
    0.52
     hâte
    0.51
     mourut
    0.51
    Activations Density 1.340%

    No Known Activations