© 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. 19154
    Prev
    Next
    INDEX
    Explanations

    Website code and data

    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
     dry
    -0.33
    utow
    -0.28
     blind
    -0.27
    dry
    -0.27
    åĨ²
    -0.27
    uted
    -0.27
    oute
    -0.26
    blind
    -0.26
    besch
    -0.26
    å¹²
    -0.25
    POSITIVE LOGITS
     nal
    0.27
    nev
    0.25
    ndata
    0.25
    ivas
    0.25
    ADA
    0.25
    nard
    0.24
    _Callback
    0.24
    _il
    0.24
    SOR
    0.24
    yat
    0.24
    Activations Density 0.010%

    No Known Activations