© 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. 39097
    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
    æĥħ
    -0.28
    æ£ĺ
    -0.26
    代
    -0.25
    åı£æ°´
    -0.25
    éĻį
    -0.25
    å¿ĥ
    -0.25
    ä¼Ļ
    -0.24
    åĸĬ
    -0.24
    bad
    -0.24
    ro
    -0.24
    POSITIVE LOGITS
    å¹³åĿĩæ°´å¹³
    0.28
    avian
    0.26
    elfth
    0.25
    çµ·
    0.24
    趺
    0.24
     attain
    0.24
    _SUITE
    0.24
     Orn
    0.24
    ushman
    0.24
    Dealer
    0.24
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.