© 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. 40348
    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
    utow
    -0.31
    åį¸
    -0.31
     Ủ
    -0.25
     hå
    -0.25
    UDA
    -0.25
     hashed
    -0.24
    ALER
    -0.24
    .bz
    -0.23
    éŁ³
    -0.23
    alion
    -0.23
    POSITIVE LOGITS
     Mid
    0.29
     Fle
    0.28
     Large
    0.28
     Industries
    0.27
     Dre
    0.26
    tracts
    0.26
     Fro
    0.26
     Early
    0.26
     End
    0.26
    missions
    0.26
    Activations Density 30.763%

    No Known Activations

    This feature has no known activations.