© 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. 31-TRANSCODER-HP
    7. 19180
    Prev
    Next
    INDEX
    Explanations

    HIV and cells

    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
    papers
    -0.28
    èĪģ
    -0.26
    ograms
    -0.26
    å¼Ģå¹ķ
    -0.26
    Editing
    -0.26
    stoff
    -0.26
     documenting
    -0.25
    lide
    -0.25
    mapped
    -0.25
    rique
    -0.25
    POSITIVE LOGITS
     attachment
    0.32
    伺
    0.31
     forward
    0.28
     Inc
    0.28
    enton
    0.26
     attach
    0.26
    dc
    0.25
    ail
    0.25
     entry
    0.25
     capital
    0.25
    Activations Density 0.224%

    No Known Activations