© 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. 87844
    Prev
    Next
    INDEX
    Explanations

    Statistics and percentages

    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
     creative
    -0.28
    aine
    -0.27
     Recreation
    -0.25
    .esp
    -0.25
    (or
    -0.25
    æ¡Ī
    -0.24
    并
    -0.24
    proper
    -0.23
    .Num
    -0.23
    act
    -0.23
    POSITIVE LOGITS
    åĴĮ
    0.35
    åĪĨåĪ«æĺ¯
    0.31
    ãĢģ
    0.30
    ãĢģãĢĮ
    0.29
    åĪĨåĪ¥
    0.28
    以åıĬ
    0.28
    åĴĮæĶ¯æĮģ
    0.28
    -,
    0.27
     sides
    0.27
    çŃīè¡Įä¸ļ
    0.27
    Activations Density 0.005%

    No Known Activations