© 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. 22-TRANSCODER-HP
    7. 130395
    Prev
    Next
    INDEX
    Explanations

    medical/biological texts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .GetFiles
    -0.25
    aab
    -0.25
    جÙĬ
    -0.25
     those
    -0.24
     ramp
    -0.23
    %%*/
    -0.23
    .Me
    -0.23
    held
    -0.23
    egin
    -0.23
    arty
    -0.23
    POSITIVE LOGITS
    åĺĮ
    0.27
    å¼¹
    0.27
    ç¬ijè¯Ŀ
    0.26
    ç¬ijäºĨ
    0.25
    oyal
    0.25
    lig
    0.25
    èĥĨ
    0.25
    馥
    0.24
    æµģ
    0.24
    èħº
    0.24
    Activations Density 0.019%

    No Known Activations