© 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. 33-TRANSCODER-HP
    7. 54531
    Prev
    Next
    INDEX
    Explanations

    urban environments

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.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±Ĭæ¯ķä¸ļçĶŁ
    -0.29
    mö
    -0.28
     gained
    -0.27
    )||(
    -0.27
    责任ç¼ĸè¾ij
    -0.26
    mention
    -0.26
    太åŃIJ
    -0.25
    acock
    -0.25
    nard
    -0.25
    äºĮæīĭ车
    -0.25
    POSITIVE LOGITS
    yl
    0.30
    sk
    0.29
     mot
    0.28
    hi
    0.26
    æ¼Ķ
    0.26
    ter
    0.26
    éļĶ
    0.26
    å§Ķç»Ħç»ĩ
    0.25
    si
    0.25
     cl
    0.25
    Activations Density 0.016%

    No Known Activations