© 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. 35-TRANSCODER-HP
    7. 40083
    Prev
    Next
    INDEX
    Explanations

    code and paths

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Greens
    -0.26
     incoming
    -0.25
     cuck
    -0.24
     rewind
    -0.24
     laundry
    -0.24
     _:
    -0.24
     intra
    -0.23
     disconnect
    -0.23
     nasty
    -0.23
     proceed
    -0.23
    POSITIVE LOGITS
    ectl
    0.31
    bilt
    0.31
    olo
    0.29
     Trev
    0.29
    loo
    0.28
    bu
    0.28
    ä¸Ģ種
    0.28
    doll
    0.27
    ouri
    0.27
    mom
    0.27
    Activations Density 0.042%

    No Known Activations