© 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. 63862
    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
    å¤ĸåĽ´
    -0.28
    鬣
    -0.25
    ëį©
    -0.25
     sens
    -0.25
    WithContext
    -0.25
    æ²¾
    -0.24
    -drop
    -0.24
    éªļ
    -0.24
    å¤ĸéĥ¨
    -0.24
    轶
    -0.24
    POSITIVE LOGITS
    ][:
    0.30
    åħ¬ç¤º
    0.29
    acre
    0.29
     finally
    0.28
    å¤į
    0.27
    ç»Īäºİ
    0.26
    ãĤ³ãĥ¬
    0.26
    emark
    0.25
     [:
    0.25
    finally
    0.24
    Activations Density 1.049%

    No Known Activations

    This feature has no known activations.