© 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. 32-TRANSCODER-HP
    7. 42081
    Prev
    Next
    INDEX
    Explanations

    and

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rends
    -0.27
    -inline
    -0.27
    adian
    -0.26
    exual
    -0.26
    integral
    -0.24
    andy
    -0.24
    ä¿Ŀ
    -0.24
    hower
    -0.24
    styled
    -0.24
     Shiite
    -0.24
    POSITIVE LOGITS
     da
    0.33
     atmosphere
    0.32
     depth
    0.31
     ba
    0.30
    ba
    0.29
     ac
    0.28
    ya
    0.28
    da
    0.28
     ag
    0.28
     datum
    0.27
    Activations Density 0.001%

    No Known Activations