© 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. 14-TRANSCODER-HP
    7. 41744
    Prev
    Next
    INDEX
    Explanations

    Data processing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    kf
    -0.29
    foil
    -0.28
    ħį
    -0.26
    éľ
    -0.25
    åıĪæĺ¯
    -0.25
    luck
    -0.25
    èĶ¬
    -0.25
    éħįå¤ĩ
    -0.25
    loom
    -0.25
    ä¸įæĸ¹ä¾¿
    -0.24
    POSITIVE LOGITS
    INDER
    0.28
    ÙĪØ´
    0.28
    held
    0.27
    beat
    0.26
    ye
    0.26
    禹
    0.26
    æĮ½æķij
    0.26
    亮çĤ¹
    0.25
    iosa
    0.25
    (shape
    0.24
    Activations Density 1.046%

    No Known Activations