© 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. 20-TRANSCODER-HP
    7. 108020
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æł¸
    -0.30
    xcc
    -0.27
    çĿĢ
    -0.26
     ret
    -0.25
    âĪĴ
    -0.24
    ographic
    -0.24
    æĴ¤ç¦»
    -0.24
    æ¢Ĺ
    -0.23
    èijĹ
    -0.23
    éĹ´çļĦ
    -0.23
    POSITIVE LOGITS
    ackle
    0.29
    以人为
    0.29
    èĽ°
    0.27
    èľĩ
    0.27
    åħĪ导
    0.27
     Drum
    0.25
    åı²ä¸ĬæľĢ
    0.25
    ONGO
    0.25
    idl
    0.25
     slim
    0.25
    Activations Density 32.801%

    No Known Activations

    This feature has no known activations.