© 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. 142521
    Prev
    Next
    INDEX
    Explanations

    Abbreviations and initial letters

    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
    egin
    -0.28
     supervise
    -0.28
    IDA
    -0.25
    åı³è¾¹
    -0.24
    -placement
    -0.24
    ekt
    -0.24
     setC
    -0.24
    没æľīå¿ħè¦ģ
    -0.23
    oge
    -0.23
    bay
    -0.23
    POSITIVE LOGITS
    jong
    0.30
    ä¸Ģå®ļçļĦ
    0.25
     Fuel
    0.25
    anth
    0.25
    lagen
    0.23
    ãĥĩãĤ£ãĥ³ãĤ°
    0.23
    ews
    0.23
    erals
    0.23
    apped
    0.23
    ressed
    0.22
    Activations Density 17.262%

    No Known Activations