© 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. 87248
    Prev
    Next
    INDEX
    Explanations

    in

    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
    vere
    -0.29
    sten
    -0.25
     Squadron
    -0.23
    []↵
    -0.23
    Wo
    -0.23
    çĮ®
    -0.23
    äº®çľ¼
    -0.23
     Ow
    -0.22
     constrain
    -0.22
    æĸij
    -0.22
    POSITIVE LOGITS
    客æ°Ķ
    0.33
    å¾®å¦Ļ
    0.26
    侥幸
    0.25
    azing
    0.24
    主宰
    0.24
     stacks
    0.23
    ime
    0.23
    storm
    0.23
    aster
    0.23
     separates
    0.23
    Activations Density 0.008%

    No Known Activations