© 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. 19120
    Prev
    Next
    INDEX
    Explanations

    technical and scientific text

    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
    mus
    -0.28
    roe
    -0.27
     Dess
    -0.27
    rosse
    -0.27
    ogan
    -0.27
     Ingen
    -0.26
    ophage
    -0.25
    paginate
    -0.25
    indle
    -0.25
    robe
    -0.24
    POSITIVE LOGITS
    ysi
    0.30
    ency
    0.30
    ative
    0.29
     basis
    0.29
    ativ
    0.28
    basis
    0.27
    ãĤģãģ¦
    0.27
     ed
    0.26
    ativa
    0.26
    centage
    0.26
    Activations Density 0.004%

    No Known Activations