© 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. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 108674
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    .createClass
    -0.29
    unda
    -0.28
    icine
    -0.28
    åķĨçĶ¨
    -0.25
    çıŃ
    -0.25
    uggling
    -0.25
    .sap
    -0.25
    stance
    -0.25
    ãĥ¬ãĤ¹ãĥĪ
    -0.25
    od
    -0.24
    POSITIVE LOGITS
    Behaviour
    0.29
     authorised
    0.26
     behaviours
    0.26
    宣è¨Ģ
    0.25
     Killed
    0.25
    çĥĺ
    0.25
     Behaviour
    0.25
     adapted
    0.24
    åĬ¨éĿĻ
    0.24
     behaviour
    0.24
    Activations Density 12.064%

    No Known Activations

    This feature has no known activations.