© 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. 54098
    Prev
    Next
    INDEX
    Explanations

    0 or 5

    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
    otron
    -0.32
     licence
    -0.26
    å¥Ķ
    -0.26
     Buyer
    -0.25
    anzi
    -0.24
     Theft
    -0.24
     roommate
    -0.24
    åĴ¨
    -0.24
    Į¨
    -0.24
    åŀĵ
    -0.24
    POSITIVE LOGITS
     mc
    0.36
    麦
    0.35
    mc
    0.35
     mcc
    0.33
     power
    0.31
     MC
    0.30
    nip
    0.30
    power
    0.30
     CF
    0.29
    pow
    0.28
    Activations Density 0.012%

    No Known Activations