© 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. 30-TRANSCODER-HP
    7. 96800
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çłĴ
    -0.29
    äºĭ件
    -0.26
     Events
    -0.26
     events
    -0.26
     Attach
    -0.25
     defer
    -0.25
     student
    -0.25
    .Payload
    -0.25
     event
    -0.25
    ãĥ¡ãĥ³ãĥĪ
    -0.25
    POSITIVE LOGITS
    sy
    0.29
    ERE
    0.27
    clipse
    0.26
    å¼ŁåħĦ
    0.26
    ornings
    0.26
    /pm
    0.26
    _sy
    0.25
    èĻ½
    0.25
    yx
    0.25
    èles
    0.25
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.