© 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. 86950
    Prev
    Next
    INDEX
    Explanations

    measurements and numbers

    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
    Negative Logits
     gef
    -0.26
    ä¸Ģæĸ¹
    -0.26
    -win
    -0.26
    DK
    -0.25
    å°±ä¸įèĥ½
    -0.24
    /aws
    -0.24
    éĶ·
    -0.23
    ä¸ĢåĪ»
    -0.23
    ä¸ĢæµģçļĦ
    -0.23
    æĬµ
    -0.23
    POSITIVE LOGITS
    .AddParameter
    0.31
    çľŁäºº
    0.26
    ´Ģ
    0.26
    vet
    0.25
    middlewares
    0.24
     ìĦ¸
    0.24
    insky
    0.24
    toArray
    0.23
    à¸ģา
    0.23
    ä½ķ
    0.23
    Activations Density 0.010%

    No Known Activations