© 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. 18029
    Prev
    Next
    INDEX
    Explanations

    decomposing matter

    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
    ported
    -0.28
    iler
    -0.26
    nels
    -0.25
     nail
    -0.25
    posit
    -0.25
     Bernstein
    -0.25
    ammed
    -0.24
    çī¹èī²
    -0.24
    Opens
    -0.24
    Open
    -0.24
    POSITIVE LOGITS
     breaker
    0.30
    _DEC
    0.30
    _dec
    0.29
     BREAK
    0.29
    DEC
    0.29
    cae
    0.29
     Breaking
    0.28
    磮
    0.28
     breaking
    0.27
     Break
    0.27
    Activations Density 1.343%

    No Known Activations