© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 3003
    Prev
    Next
    INDEX
    Explanations

    the terms "LED", "heuristic", "imperialistic" and words ending in "istic"

    oai_token-act-pair · gemini-2.0-flash

    Technical/abstract discussions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     
    -1.38
     (
    -1.09
     "
    -1.07
     “
    -1.07
    es
    -1.07
    ,
    -1.05
      
    -1.05
     a
    -0.99
     or
    -0.96
     I
    -0.96
    POSITIVE LOGITS
     ainfi
    1.80
     auffi
    1.77
     feroit
    1.74
     Efq
    1.73
     AssemblyCulture
    1.73
     étoient
    1.70
     Jefus
    1.68
     pouvoit
    1.67
     auroit
    1.67
     Monfieur
    1.66
    Activations Density 2.970%

    No Known Activations