© 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. 22-TRANSCODER-HP
    7. 87097
    Prev
    Next
    INDEX
    Explanations

    Code, legal, and scientific texts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tection
    -0.26
     ourselves
    -0.25
    æĸ½
    -0.25
    enda
    -0.24
    amar
    -0.24
    常ãģ«
    -0.23
    á»ģ
    -0.23
     always
    -0.23
    åĩĢ
    -0.23
    åıĹåĪ°
    -0.23
    POSITIVE LOGITS
    åĪĿ次
    0.28
    é¦ĸ次
    0.28
    (tm
    0.25
    好è¯Ħ
    0.25
    utive
    0.25
     entirely
    0.24
    ä¸ī代
    0.24
    è¦ģçŁ¥éģĵ
    0.24
    æľīæ¯Ĵ
    0.24
    è°Ĥ
    0.24
    Activations Density 0.001%

    No Known Activations