© 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. 31-TRANSCODER-HP
    7. 18539
    Prev
    Next
    INDEX
    Explanations

    Van

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĪIJå°±
    -0.29
    âĩIJ
    -0.26
    ä¸Ĭç½ij
    -0.25
     BIND
    -0.25
    æĭĪ
    -0.24
    	Optional
    -0.24
    天èµĭ
    -0.24
    èĢģåħµ
    -0.23
    -config
    -0.23
    yan
    -0.23
    POSITIVE LOGITS
    ãĤ¹ãĥŀ
    0.28
     Parties
    0.28
    /stdc
    0.28
    è¿ĩæĿ¥
    0.28
    endar
    0.27
    çģ«çĥ§
    0.27
    æľ¬æĬ¥
    0.25
    cope
    0.25
    åIJĦæł·
    0.24
    GREE
    0.24
    Activations Density 0.004%

    No Known Activations