© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 96805
    Prev
    Next
    INDEX
    Explanations

    Natural vs something else

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     natural
    -0.28
    aris
    -0.28
    æ´»
    -0.27
    æĬ¥çº¸
    -0.26
     programme
    -0.26
    FINITE
    -0.25
    errorCode
    -0.25
    çIJĨè«ĸ
    -0.25
    aira
    -0.25
     dining
    -0.24
    POSITIVE LOGITS
    工夫
    0.27
    ä¸ĢæĹ¥
    0.26
     ob
    0.26
    bourg
    0.25
    esz
    0.24
     Gather
    0.24
    ).__
    0.24
     Haz
    0.24
    佩
    0.24
    è§ĦåĪĴ建设
    0.23
    Activations Density 0.782%

    No Known Activations