© 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. 18360
    Prev
    Next
    INDEX
    Explanations

    casual conversation 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
     UR
    -0.28
    ei
    -0.26
    åºĸ
    -0.25
    oton
    -0.25
    avad
    -0.25
    ierre
    -0.25
    æĭī
    -0.25
    laus
    -0.24
     Its
    -0.24
    Its
    -0.24
    POSITIVE LOGITS
     forward
    0.33
    forward
    0.29
     forwards
    0.28
    zell
    0.28
     public
    0.27
    -forward
    0.27
    resent
    0.26
    ublic
    0.26
    èµ°åIJij
    0.26
    åķĨä¼ļ
    0.26
    Activations Density 0.818%

    No Known Activations