© 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. 97894
    Prev
    Next
    INDEX
    Explanations

    Contrasting ideas/options

    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ä¸įæĺ¯
    -0.30
    street
    -0.29
    巨人
    -0.27
    并éĿŀ
    -0.27
    elmet
    -0.27
    éĥ½ä¸įæĺ¯
    -0.26
    å¦Ĥæŀľä¸įæĺ¯
    -0.26
     ↵        ↵
    -0.25
    ;++
    -0.25
    both
    -0.25
    POSITIVE LOGITS
    çļĦè¿ĽæŃ¥
    0.28
    è¿ĽæŃ¥
    0.28
    éĻ¬
    0.27
    个åĪ«
    0.27
    æľī个
    0.25
     compound
    0.24
    outines
    0.24
    rio
    0.24
     scanned
    0.24
    çļĦåIJĪä½ľ
    0.24
    Activations Density 0.006%

    No Known Activations