© 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. 39365
    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
    emploi
    -0.28
    æµĵåİļ
    -0.27
    Å¡e
    -0.27
    æ¿Ģåıij
    -0.27
    强çĥĪçļĦ
    -0.26
     Rentals
    -0.26
    浸
    -0.26
    бан
    -0.26
    prove
    -0.25
     excited
    -0.25
    POSITIVE LOGITS
     down
    0.28
    inger
    0.26
     Down
    0.26
     enter
    0.26
    ures
    0.25
     [*
    0.25
     win
    0.25
     bu
    0.25
     (âĪ
    0.25
    åĬ¡
    0.24
    Activations Density 3.099%

    No Known Activations

    This feature has no known activations.