© 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. 12-TRANSCODER-HP
    7. 24100
    Prev
    Next
    INDEX
    Explanations

    labels

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    è§Ħå®ļçļĦ
    -0.31
     qualify
    -0.30
    ä¸ĢçĶŁ
    -0.29
    /stdc
    -0.28
     bande
    -0.28
     fest
    -0.27
    mos
    -0.27
     permitted
    -0.27
    èĬĤæĹ¥
    -0.27
    âm
    -0.26
    POSITIVE LOGITS
    anj
    0.32
    omm
    0.27
     dinosaur
    0.26
     planetary
    0.25
     tiếng
    0.25
    è·¯çģ¯
    0.24
    æĭIJ
    0.24
    ä»Ĺ
    0.24
     Citadel
    0.24
    éªĢ
    0.24
    Activations Density 0.000%

    No Known Activations