© 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. 108775
    Prev
    Next
    INDEX
    Explanations

    Images and attachments

    np_max-act · gemini-2.0-flash
    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
    ampo
    -0.26
    没éĹ®é¢ĺ
    -0.26
    rô
    -0.24
    ways
    -0.24
    ìĸij
    -0.24
    âĪħ
    -0.24
     finite
    -0.23
    èĸ¤
    -0.23
    æľīéĻIJ
    -0.23
    uire
    -0.23
    POSITIVE LOGITS
    ivan
    0.27
    çĤ¹çĩĥ
    0.25
     fond
    0.25
    .INSTANCE
    0.25
    AIN
    0.24
    ipt
    0.24
     contamin
    0.23
    izard
    0.23
    PORT
    0.23
    Separated
    0.23
    Activations Density 1.843%

    No Known Activations