© 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. 19483
    Prev
    Next
    INDEX
    Explanations

    Code and license text

    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
    éĻ¢
    -0.33
    encer
    -0.32
    open
    -0.29
     Institutions
    -0.28
     Open
    -0.28
    .Open
    -0.28
    Open
    -0.26
     map
    -0.26
     open
    -0.26
    map
    -0.25
    POSITIVE LOGITS
    è¡į
    0.34
    ä¹ī
    0.26
    大äºĭ
    0.25
    åĩ»
    0.24
    urar
    0.24
    æĢħ
    0.24
    éģ´
    0.24
     Quint
    0.24
     gens
    0.24
    .si
    0.23
    Activations Density 0.007%

    No Known Activations