© 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. 25-TRANSCODER-HP
    7. 109165
    Prev
    Next
    INDEX
    Explanations

    ircuit

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    er
    -0.28
    pected
    -0.27
    çĿģå¼Ģ
    -0.26
    OND
    -0.25
    athers
    -0.25
    iders
    -0.23
    kj
    -0.23
    etch
    -0.23
     Airways
    -0.23
    ather
    -0.23
    POSITIVE LOGITS
     hookup
    0.24
    å°ļæľª
    0.24
    een
    0.24
    umes
    0.24
    ickle
    0.23
    uits
    0.23
    REAM
    0.23
    é£İåħī
    0.23
    üm
    0.23
    .EVENT
    0.22
    Activations Density 0.636%

    No Known Activations