© 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. 18645
    Prev
    Next
    INDEX
    Explanations

    technical or code snippets

    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
    ader
    -0.29
     Straw
    -0.28
     Desmond
    -0.28
    cea
    -0.27
    emes
    -0.27
    orns
    -0.26
    UED
    -0.26
    ç¯Ģ
    -0.25
    声
    -0.25
    -paper
    -0.25
    POSITIVE LOGITS
    Cr
    0.37
     Mn
    0.36
     Cr
    0.34
     cr
    0.33
    Mo
    0.32
     Ni
    0.31
    Si
    0.30
    Ni
    0.29
     Mo
    0.29
     nich
    0.29
    Activations Density 4.541%

    No Known Activations