© 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. 22-TRANSCODER-HP
    7. 98503
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     EFFECT
    -0.26
     Alic
    -0.25
    æıIJ
    -0.25
     Minds
    -0.24
    ONY
    -0.24
    éĻ¬
    -0.24
    gradable
    -0.23
    &q
    -0.23
    继
    -0.23
     Attach
    -0.23
    POSITIVE LOGITS
    çľĭä»ĸ
    0.30
    翻身
    0.26
    trees
    0.25
    ÑģÑĤо
    0.25
    éº»æľ¨
    0.25
    泯
    0.24
     pane
    0.24
    åŃĺæ´»
    0.24
    çĶľèľľ
    0.24
    çļĦçĶŁåij½
    0.23
    Activations Density 0.023%

    No Known Activations

    This feature has no known activations.