© 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. 32-TRANSCODER-HP
    7. 43301
    Prev
    Next
    INDEX
    Explanations

    grey/gray

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .Decode
    -0.28
    .au
    -0.27
    ars
    -0.27
    y
    -0.26
    pires
    -0.26
    åļ¼
    -0.25
    .onStart
    -0.25
    ients
    -0.25
    uggested
    -0.25
    ivating
    -0.25
    POSITIVE LOGITS
     color
    0.52
    -color
    0.49
    color
    0.49
    èī²
    0.47
     colour
    0.45
    Color
    0.44
    é¢ľèī²
    0.43
    colour
    0.43
    .color
    0.41
    -colored
    0.41
    Activations Density 0.004%

    No Known Activations