© 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. 111178
    Prev
    Next
    INDEX
    Explanations

    Locations

    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
     wisely
    -0.27
    udden
    -0.24
    âį¨
    -0.24
    ignon
    -0.23
    yst
    -0.23
    ackson
    -0.23
    Portal
    -0.23
    uhl
    -0.23
    dyn
    -0.23
    igen
    -0.23
    POSITIVE LOGITS
    åĪ°è¿ĻéĩĮ
    0.30
    asters
    0.29
    Ĥ¨
    0.26
    bers
    0.26
     Lauderdale
    0.25
    isma
    0.25
    ENCES
    0.25
    AY
    0.24
    èĪĴ
    0.24
    æ°Ķåij³
    0.24
    Activations Density 0.047%

    No Known Activations