© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 24-TRANSCODER-HP
    7. 149432
    Prev
    Next
    INDEX
    Explanations

    Construction and structure

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    鬣
    -0.28
    ertain
    -0.28
     structure
    -0.27
    ä¹łæĥ¯äºĨ
    -0.26
    hunt
    -0.26
    çµIJæ§ĭ
    -0.26
    èł¹
    -0.26
     yourselves
    -0.25
    heels
    -0.25
    ç»ĵæŀĦ
    -0.24
    POSITIVE LOGITS
    ivist
    0.31
    å¸Ŀ
    0.28
    IBLE
    0.27
    ors
    0.26
    ible
    0.26
    acular
    0.25
    aut
    0.24
    ORS
    0.24
    ilio
    0.24
     criticism
    0.24
    Activations Density 1.212%

    No Known Activations