© 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. 19921
    Prev
    Next
    INDEX
    Explanations

    eligibility conditions

    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
    Negative Logits
    imary
    -0.29
    èµŁ
    -0.27
     attendee
    -0.27
    è§ij
    -0.26
     explorer
    -0.26
    å°ıåŃIJ
    -0.26
    åĮħ
    -0.25
     colonies
    -0.24
    pins
    -0.24
    hop
    -0.24
    POSITIVE LOGITS
     someone
    0.34
     flowing
    0.31
    ç»´æĮģ
    0.30
     running
    0.30
     conducting
    0.29
    someone
    0.29
     performing
    0.28
     working
    0.28
     operating
    0.28
     participating
    0.28
    Activations Density 0.287%

    No Known Activations