© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 98404
    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
    Not in Any Lists

    No Comments

    Negative Logits
     alone
    -0.26
    è¿ĻéĩĮ
    -0.25
    å°±è¦ģ
    -0.24
    IPC
    -0.24
    astery
    -0.24
     languages
    -0.23
     luck
    -0.23
     Dude
    -0.23
    PET
    -0.23
    æľ¬èº«
    -0.22
    POSITIVE LOGITS
     Stall
    0.27
     FAR
    0.26
     dr
    0.26
    mares
    0.25
    uchs
    0.24
    -born
    0.24
    fal
    0.24
    rates
    0.24
    æĢ»å±Ģ
    0.24
    è£Ĥ
    0.23
    Activations Density 0.008%

    No Known Activations

    This feature has no known activations.