© 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. 41125
    Prev
    Next
    INDEX
    Explanations

    domain

    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
    Not in Any Lists

    No Comments

    Negative Logits
    vit
    -0.27
    afil
    -0.26
    )V
    -0.26
    nip
    -0.26
     Cold
    -0.26
     Forever
    -0.25
    åĨ³
    -0.25
    vat
    -0.25
     recalled
    -0.25
     VARIANT
    -0.25
    POSITIVE LOGITS
    æİ©
    0.28
     degrees
    0.28
    PAR
    0.25
     agent
    0.25
    ellan
    0.25
    USES
    0.25
     PAR
    0.25
    éĹ¾
    0.24
    åŁºäºİ
    0.24
    degrees
    0.24
    Activations Density 0.003%

    No Known Activations