© 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. 32-TRANSCODER-HP
    7. 40835
    Prev
    Next
    INDEX
    Explanations

    specializes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    rapped
    -0.30
    å¼Ģ
    -0.27
    å°ij
    -0.26
     anyhow
    -0.25
    çĶ¨éĢĶ
    -0.25
    æł¡
    -0.25
    ä¸ī项
    -0.24
    once
    -0.24
    ä¹Łæ¯Ķè¾ĥ
    -0.24
    uppe
    -0.24
    POSITIVE LOGITS
    äºİ
    0.27
    .transitions
    0.27
    peat
    0.27
    çİ¢
    0.26
     paycheck
    0.25
    æŀ°
    0.25
     aboard
    0.25
    ipop
    0.25
    [âĢ¦
    0.24
    pest
    0.24
    Activations Density 0.003%

    No Known Activations