© 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. 18535
    Prev
    Next
    INDEX
    Explanations

    code and database questions

    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
     MC
    -0.27
    KL
    -0.26
    KO
    -0.26
     setDescription
    -0.26
    кл
    -0.25
    VML
    -0.25
    KM
    -0.25
    ä¾Ħ
    -0.24
    .grade
    -0.24
    grade
    -0.24
    POSITIVE LOGITS
    ya
    0.29
     dep
    0.26
     perish
    0.25
    æ®·
    0.25
    ategoria
    0.24
    å·´
    0.24
    amba
    0.24
    pora
    0.24
    å¦ĸ
    0.24
    olars
    0.23
    Activations Density 3.601%

    No Known Activations