© 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. 20-TRANSCODER-HP
    7. 108748
    Prev
    Next
    INDEX
    Explanations

    instructions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    è°
    -0.30
    (LL
    -0.28
    uous
    -0.26
     loaders
    -0.26
    _ask
    -0.25
    placeholder
    -0.25
    llx
    -0.25
    (PC
    -0.25
     prompts
    -0.24
     settled
    -0.24
    POSITIVE LOGITS
    itaire
    0.26
     outnumber
    0.26
    ä»ħä»ħæĺ¯
    0.26
     indeed
    0.26
    l
    0.26
     chief
    0.25
    ov
    0.25
    OTP
    0.25
    Datum
    0.25
    çIJĨçĶ±
    0.25
    Activations Density 54.511%

    No Known Activations