© 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. 30-TRANSCODER-HP
    7. 94940
    Prev
    Next
    INDEX
    Explanations

    Operating systems/coding context

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    tober
    -0.28
     prot
    -0.27
    oshi
    -0.26
     discharge
    -0.25
    zos
    -0.25
    é£Łåĵģ
    -0.24
    rema
    -0.24
     smallest
    -0.23
    asty
    -0.23
    !--
    -0.23
    POSITIVE LOGITS
    ergic
    0.28
    åħ¬è¯ī
    0.27
    svc
    0.27
     legalized
    0.26
    bullet
    0.26
     Pix
    0.25
    .inv
    0.24
    caster
    0.24
    abouts
    0.24
    Pix
    0.24
    Activations Density 0.884%

    No Known Activations