© 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. 99162
    Prev
    Next
    INDEX
    Explanations

    sediment

    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
    sert
    -0.27
    sing
    -0.27
    تج
    -0.27
     favour
    -0.26
     favourites
    -0.25
    agus
    -0.25
    åij½
    -0.24
    çijŁ
    -0.23
    abstract
    -0.23
    ä¸Ģå®ļ
    -0.23
    POSITIVE LOGITS
     analy
    0.29
    /her
    0.28
    ä¸į足
    0.27
    iotic
    0.24
    _backward
    0.24
     backward
    0.24
    ence
    0.24
     kicking
    0.23
     base
    0.23
    ependency
    0.23
    Activations Density 0.155%

    No Known Activations