© 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. 52971
    Prev
    Next
    INDEX
    Explanations

    Aaron

    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
    ter
    -0.27
    ROL
    -0.27
    rolled
    -0.27
    xAE
    -0.26
    (pass
    -0.26
    xB
    -0.25
    icine
    -0.24
    pend
    -0.23
    ffer
    -0.23
    Rol
    -0.23
    POSITIVE LOGITS
    骨
    0.29
    ä½ķ
    0.29
    Val
    0.26
     pig
    0.25
     Val
    0.25
    åIJ¦
    0.25
     Disconnect
    0.25
    igi
    0.24
    éĺĢ
    0.24
    æѦ
    0.24
    Activations Density 0.005%

    No Known Activations