© 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. 32-TRANSCODER-HP
    7. 41041
    Prev
    Next
    INDEX
    Explanations

    spinal references

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    rh
    -0.28
     Nb
    -0.27
    ait
    -0.26
    æĭ©
    -0.26
    elsius
    -0.26
    WC
    -0.26
    ose
    -0.26
    .Fl
    -0.24
    åºĹéĩĮ
    -0.24
     WC
    -0.24
    POSITIVE LOGITS
    dish
    0.28
    å¾·å°Ķ
    0.27
    odian
    0.26
    aneous
    0.26
    arLayout
    0.26
    çķĮå®ļ
    0.26
    sweet
    0.26
    uster
    0.26
    yal
    0.25
    dì
    0.25
    Activations Density 0.002%

    No Known Activations