© 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. 112339
    Prev
    Next
    INDEX
    Explanations

    interchangeable

    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
    çĥĻ
    -0.28
    less
    -0.27
    笼
    -0.27
    éĩĮç¨ĭç¢ij
    -0.26
     __________________↵↵
    -0.25
    èĩªåıij
    -0.25
    åĪ©äºļ
    -0.25
    race
    -0.24
    çĽĽå®´
    -0.24
    æĢĢçĿĢ
    -0.24
    POSITIVE LOGITS
    (EXPR
    0.25
     garbage
    0.24
    ERE
    0.24
    -flex
    0.24
     Faster
    0.23
    ä¸įäºĪ
    0.23
    ely
    0.23
    TestClass
    0.23
    (NULL
    0.22
     grants
    0.22
    Activations Density 0.104%

    No Known Activations