© 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. 31-TRANSCODER-HP
    7. 19618
    Prev
    Next
    INDEX
    Explanations

    micro

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ably
    -0.27
    пÑĥÑģÑĤ
    -0.27
    çķ´
    -0.26
     simp
    -0.26
     Fant
    -0.25
     atop
    -0.25
     cylinders
    -0.25
     coun
    -0.25
     semi
    -0.24
     Cy
    -0.24
    POSITIVE LOGITS
    s
    0.64
    sar
    0.40
    sut
    0.35
    sand
    0.34
    sip
    0.33
    scape
    0.32
    sse
    0.32
    sx
    0.32
    sheets
    0.30
    erna
    0.29
    Activations Density 0.010%

    No Known Activations