© 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. 31-TRANSCODER-HP
    7. 19612
    Prev
    Next
    INDEX
    Explanations

    code or programming snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cent
    -0.27
    éĨĮ
    -0.26
    ãģ®ãģĵãģ¨
    -0.26
    èĸľ
    -0.26
    (low
    -0.25
    çĭ¬
    -0.25
    éĹŃ
    -0.25
    åıĺäºĨ
    -0.25
    çį¨
    -0.24
     quarantine
    -0.24
    POSITIVE LOGITS
     recre
    0.28
    strap
    0.26
     cookie
    0.25
    ain
    0.25
    .helper
    0.25
    èµĭ
    0.25
    æĺĨèĻ«
    0.25
    unk
    0.25
    篮
    0.24
    .instrument
    0.24
    Activations Density 1.081%

    No Known Activations