© 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. 40162
    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
    ortho
    -0.27
    对åѦçĶŁ
    -0.27
    æĥħ人èĬĤ
    -0.26
    æĭ¼åij½
    -0.25
    éĩįä¸Ń
    -0.25
    ä¸ĩ象
    -0.25
    åħ¨å¹´
    -0.25
    agina
    -0.25
    pag
    -0.25
    alez
    -0.24
    POSITIVE LOGITS
    uron
    0.29
     vict
    0.28
     uy
    0.26
    âĢ¦↵↵↵
    0.25
    âĢ¦"
    0.25
     unlocked
    0.24
    sm
    0.24
     fascinated
    0.23
     converge
    0.23
     mell
    0.23
    Activations Density 0.016%

    No Known Activations

    This feature has no known activations.