© 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. 18485
    Prev
    Next
    INDEX
    Explanations

    Car models and specs

    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
    追
    -0.30
     horns
    -0.29
     pumps
    -0.27
     Oper
    -0.27
     raided
    -0.26
    ä¸ĭ
    -0.26
    åĽ¢
    -0.26
    ĵį
    -0.25
    units
    -0.24
     sidel
    -0.24
    POSITIVE LOGITS
    вин
    0.31
     grade
    0.30
     rated
    0.30
     grading
    0.29
     Rated
    0.28
    身æĿIJ
    0.28
     burner
    0.27
     graded
    0.26
     zip
    0.26
    zip
    0.26
    Activations Density 0.110%

    No Known Activations