© 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. 39080
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    åħīçݯ
    -0.28
    èĮ§
    -0.27
    åıĬ以ä¸Ĭ
    -0.24
    åIJ¯åĬ¨ä»ªå¼ı
    -0.23
    ä¸Ńåħ³
    -0.23
    åħīèĬĴ
    -0.23
    åīįéĢĶ
    -0.22
    ATTRIBUTE
    -0.22
    è§Ĩå¯Ł
    -0.22
    .LA
    -0.21
    POSITIVE LOGITS
    rette
    0.25
    ôle
    0.24
     Couple
    0.24
     Miner
    0.23
    files
    0.23
    ardo
    0.23
     Unauthorized
    0.22
    ichten
    0.22
    fields
    0.22
    iddy
    0.22
    Activations Density 0.016%

    No Known Activations

    This feature has no known activations.