© 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. 24-TRANSCODER-HP
    7. 149810
    Prev
    Next
    INDEX
    Explanations

    Weapons

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĹĹä¸ĭçļĦ
    -0.28
    mouth
    -0.28
    elope
    -0.26
    umen
    -0.26
    maries
    -0.25
    encers
    -0.25
     bush
    -0.24
    encing
    -0.24
    gae
    -0.24
    ances
    -0.24
    POSITIVE LOGITS
    edly
    0.28
    _neg
    0.25
     Giov
    0.25
    èĩ³æѤ
    0.24
    _NEG
    0.23
    irim
    0.23
     mnie
    0.22
    BAT
    0.22
     biome
    0.21
    幺
    0.21
    Activations Density 0.533%

    No Known Activations