© 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. 19026
    Prev
    Next
    INDEX
    Explanations

    region

    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
     gadget
    -0.28
    æ´¾
    -0.28
    mere
    -0.27
    使çĶ¨èĢħ
    -0.27
    å¼¥
    -0.26
    çļĦ人æĿ¥è¯´
    -0.26
    象
    -0.25
    è½½
    -0.25
     Gro
    -0.24
    éħįå¥Ĺ
    -0.24
    POSITIVE LOGITS
    's
    0.30
     Away
    0.28
     away
    0.27
     around
    0.27
    away
    0.27
    iaux
    0.25
     sued
    0.25
    çĭĻ
    0.25
    IAS
    0.25
     facing
    0.25
    Activations Density 0.054%

    No Known Activations