© 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. 131104
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     bowling
    -0.25
    .har
    -0.24
     Bowling
    -0.24
    -strokes
    -0.24
    urb
    -0.24
    ikh
    -0.23
    ieren
    -0.22
     billing
    -0.22
    NewLabel
    -0.22
     nett
    -0.22
    POSITIVE LOGITS
    ped
    0.30
    Initializer
    0.26
    sl
    0.24
     inher
    0.23
    é¢ĨåħĪ
    0.23
    é¢ĨåħĪçļĦ
    0.23
    rado
    0.23
     Held
    0.22
    éģĹ产
    0.22
    沦
    0.22
    Activations Density 0.029%

    No Known Activations

    This feature has no known activations.