© 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. 43786
    Prev
    Next
    INDEX
    Explanations

    say "General"

    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
    ä¸ĢèĪ¬
    -0.34
    gener
    -0.33
     generally
    -0.31
    bel
    -0.31
     general
    -0.28
    èĪ¬
    -0.27
     defaults
    -0.27
    bled
    -0.26
    ti
    -0.26
    bias
    -0.26
    POSITIVE LOGITS
    ecake
    0.25
     Potter
    0.25
    prenom
    0.24
    levelname
    0.24
     Harold
    0.23
    éĤĤ
    0.23
    esar
    0.23
     Consort
    0.23
     Crane
    0.23
    åĵ¢
    0.23
    Activations Density 0.016%

    No Known Activations