© 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. 18-TRANSCODER-HP
    7. 88683
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åħ·å¤ĩ
    -0.27
    çݯçIJĥ
    -0.26
    jal
    -0.26
    æŁIJ
    -0.26
    æ¡£
    -0.25
    æŁIJä¸Ģ
    -0.25
    æŁIJ个
    -0.24
    注éĩį
    -0.24
     adopted
    -0.24
    atures
    -0.24
    POSITIVE LOGITS
     Zip
    0.28
    HEST
    0.27
    Pocket
    0.25
    å¾Īæ¼Ĥ亮
    0.24
    PLACE
    0.24
    Dig
    0.24
    NAMESPACE
    0.24
     Dig
    0.23
    æīĵäºĨ
    0.23
    æĿ¥äºĨ
    0.23
    Activations Density 2.151%

    No Known Activations

    This feature has no known activations.