© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 33-TRANSCODER-HP
    7. 53635
    Prev
    Next
    INDEX
    Explanations

    lungs and airways

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĬ¥
    -0.32
    ded
    -0.28
    roat
    -0.28
    \Json
    -0.26
    ÃŃculo
    -0.26
    umo
    -0.25
    lay
    -0.25
    åĴ³
    -0.25
    -json
    -0.24
    _DL
    -0.24
    POSITIVE LOGITS
     wind
    0.64
    æ°Ķ
    0.63
     tr
    0.59
    é£İ
    0.59
    wind
    0.53
    tr
    0.50
    æ°£
    0.49
     Tr
    0.48
     Wind
    0.47
     air
    0.45
    Activations Density 0.303%

    No Known Activations