© 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. 33-TRANSCODER-HP
    7. 53049
    Prev
    Next
    INDEX
    Explanations

    el

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ypy
    -0.31
    çļĦæĥħ
    -0.30
     y
    -0.28
    ndata
    -0.26
    OPY
    -0.25
    -ended
    -0.25
    -blind
    -0.25
    blind
    -0.25
     attraction
    -0.23
    поÑĤ
    -0.23
    POSITIVE LOGITS
     el
    0.94
    	el
    0.73
    (el
    0.59
    -el
    0.57
    .el
    0.53
     El
    0.53
    _el
    0.52
    El
    0.49
    /el
    0.49
    el
    0.43
    Activations Density 0.080%

    No Known Activations