© 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. 19359
    Prev
    Next
    INDEX
    Explanations

    ages 70-80

    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
    åħ«å¤§
    -0.29
    ä¸į对
    -0.27
    鸡
    -0.26
    ç¼
    -0.26
    ball
    -0.25
    mant
    -0.25
    balls
    -0.25
     ksztaÅĤ
    -0.24
    packed
    -0.24
    妹
    -0.24
    POSITIVE LOGITS
     Past
    0.26
    çģ
    0.24
    ivr
    0.24
    涨
    0.24
    æĬļ
    0.24
    Atlas
    0.24
    伪
    0.23
    两次
    0.23
    ILA
    0.23
     Freder
    0.23
    Activations Density 0.004%

    No Known Activations