© 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. 130700
    Prev
    Next
    INDEX
    Explanations

    technical, code snippets

    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
    并没æľī
    -0.29
    请èģĶç³»
    -0.27
    ä½Ĩåį´
    -0.27
     liquid
    -0.26
    棵æłij
    -0.26
    并ä¸įä¼ļ
    -0.25
     pot
    -0.25
     seize
    -0.25
    Evidence
    -0.25
    éľĪ
    -0.25
    POSITIVE LOGITS
    åĴĮå¹³
    0.28
    INO
    0.28
     itself
    0.26
    ROID
    0.25
    缴èĩ³
    0.25
    ->{$
    0.24
    ROC
    0.24
     rej
    0.24
    \grid
    0.24
    redi
    0.24
    Activations Density 0.005%

    No Known Activations