© 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. 3050
    Prev
    Next
    INDEX
    Explanations

    verbs relating an action causing a result or an intermediate state

    oai_token-act-pair · gemini-2.0-flash

    led/supported

    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
     led
    -1.74
     presided
    -1.55
     caused
    -1.55
     spearheaded
    -1.53
     hosted
    -1.51
     coached
    -1.49
     financed
    -1.43
     fueled
    -1.43
     supported
    -1.41
     fuelled
    -1.41
    POSITIVE LOGITS
     help
    0.71
     support
    0.69
     offer
    0.63
    起
    0.60
    <bos>
    0.58
    attled
    0.56
     deliver
    0.54
     use
    0.54
     helping
    0.52
     provide
    0.52
    Activations Density 3.496%

    No Known Activations