© 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. 88359
    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
    æĤī
    -0.29
    charted
    -0.25
    ouncer
    -0.25
    -hit
    -0.25
    æľī人
    -0.24
    æ¶Īæģ¯ç§°
    -0.24
    plements
    -0.24
    ç¼ĸåı·
    -0.24
    ei
    -0.23
    chet
    -0.23
    POSITIVE LOGITS
    éĵŁ
    0.28
    aira
    0.26
    æ¯ı个æľĪ
    0.26
    â¼¥
    0.25
    åĽłæŀľ
    0.24
    ToWorld
    0.24
     gained
    0.24
    çµIJ
    0.24
    icut
    0.23
    )__
    0.23
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.