© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 18340
    Prev
    Next
    INDEX
    Explanations

    code and filepaths

    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
    cript
    -0.27
     arp
    -0.26
    _ar
    -0.26
    drs
    -0.25
     Arbor
    -0.25
    arat
    -0.24
    cott
    -0.24
    ownik
    -0.24
    \Annotation
    -0.24
    (ar
    -0.24
    POSITIVE LOGITS
    uli
    0.28
     ours
    0.25
    ernals
    0.25
    angan
    0.25
     iterations
    0.25
    滩
    0.24
     Determin
    0.24
     iteration
    0.24
     Garn
    0.24
    bo
    0.24
    Activations Density 0.002%

    No Known Activations