© 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. 97056
    Prev
    Next
    INDEX
    Explanations

    author

    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
    ä¸Ĭåİ»
    -0.28
    -serving
    -0.28
    serve
    -0.26
    å·²ä¹ħ
    -0.26
    å·²ä¹ħçļĦ
    -0.25
     serving
    -0.25
    terminate
    -0.25
     Twig
    -0.24
    gende
    -0.24
    Affected
    -0.24
    POSITIVE LOGITS
    apel
    0.27
    èµĦ产è¯Ħä¼°
    0.27
    etr
    0.26
    第ä¸Ģ次
    0.25
    é¦ĸ次
    0.24
     Acquisition
    0.24
     Strom
    0.24
    /o
    0.24
    æĸ°éĹ»ä¸Ńå¿ĥ
    0.24
    issan
    0.23
    Activations Density 0.019%

    No Known Activations