© 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. 149774
    Prev
    Next
    INDEX
    Explanations

    0

    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
    é«ĺä»·
    -0.26
    ROME
    -0.26
     gymn
    -0.26
    -valid
    -0.25
    .ct
    -0.25
    ::-
    -0.25
    clamation
    -0.25
    ENN
    -0.25
     Gym
    -0.24
    actors
    -0.24
    POSITIVE LOGITS
     Lust
    0.26
     dem
    0.25
    人åĿĩ
    0.24
    elden
    0.24
    ky
    0.24
     Moran
    0.24
    è¿IJ转
    0.24
     autonomous
    0.24
     eth
    0.23
    ovich
    0.23
    Activations Density 0.214%

    No Known Activations