© 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. 151155
    Prev
    Next
    INDEX
    Explanations

    selected

    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
    Not in Any Lists

    No Comments

    Negative Logits
    hook
    -0.26
    çĥŁçģ«
    -0.26
    çļĦ第ä¸Ģ个
    -0.25
    æİĪæĿĥ
    -0.24
    .TypeOf
    -0.24
    åħĥå®Ŀ
    -0.24
     repeatedly
    -0.24
    hooks
    -0.23
     første
    -0.22
    åŀĭä¼ģä¸ļ
    -0.22
    POSITIVE LOGITS
    plural
    0.31
    amedi
    0.27
    uisse
    0.26
    èĨ¦
    0.26
    lus
    0.25
    uids
    0.25
    ruits
    0.25
    UNCH
    0.25
    @d
    0.24
    INLINE
    0.24
    Activations Density 0.184%

    No Known Activations