© 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. 22293
    Prev
    Next
    INDEX
    Explanations

    force

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ASTER
    -0.31
    Tuesday
    -0.26
    å¾Īæ¸ħæ¥ļ
    -0.26
    è½®å»ĵ
    -0.26
    å¾Ģä¸Ĭ
    -0.25
     rig
    -0.25
    æĺŁæľŁ
    -0.25
     tight
    -0.24
    ention
    -0.24
     Kim
    -0.24
    POSITIVE LOGITS
    itaire
    0.30
    :id
    0.27
    å¼Ĥåľ°
    0.27
    ById
    0.26
    scratch
    0.26
    ä¾Ŀæ³ķ
    0.26
    åį¿
    0.25
     Interracial
    0.25
    æĦıè¯Ĩ
    0.24
    izz
    0.24
    Activations Density 0.005%

    No Known Activations