© 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. 33-TRANSCODER-HP
    7. 52550
    Prev
    Next
    INDEX
    Explanations

    HTML, URLs, Tranquility

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ep
    -0.30
    eg
    -0.30
    ele
    -0.28
     atom
    -0.28
    eb
    -0.28
    eat
    -0.28
    è¯Ŀ
    -0.27
     bat
    -0.27
    atom
    -0.27
    ãģ°
    -0.27
    POSITIVE LOGITS
    _procs
    0.30
    uary
    0.29
    çĶ©
    0.26
    说å¾Ĺ
    0.25
     pul
    0.25
    mers
    0.24
    omorphic
    0.24
    æĺ¯æĪijçļĦ
    0.24
     (!!
    0.24
    ometrics
    0.24
    Activations Density 0.003%

    No Known Activations