© 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. 39494
    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
     âĢ
    -0.26
     host
    -0.25
     rec
    -0.25
    éķ¿å¤§
    -0.25
     comb
    -0.25
     t
    -0.24
     av
    -0.24
    å¹²äºĭ
    -0.24
     afford
    -0.24
    des
    -0.24
    POSITIVE LOGITS
    forced
    0.27
    elastic
    0.24
    oji
    0.23
    ä¾ĭè¡Į
    0.23
     dunk
    0.23
    featured
    0.23
    æĸĹ
    0.23
    æݥ轨
    0.23
    nj
    0.23
    _drv
    0.23
    Activations Density 0.111%

    No Known Activations

    This feature has no known activations.