© 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. 88457
    Prev
    Next
    INDEX
    Explanations

    pre

    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
     whip
    -0.28
    ä¸įåıĬ
    -0.27
    ientos
    -0.25
    åIJĮçŃī
    -0.25
    REATE
    -0.24
     bench
    -0.24
    éĥijéĩį
    -0.24
     palm
    -0.23
    å¼ºåĽ½
    -0.23
    оÑģÑĤи
    -0.23
    POSITIVE LOGITS
    çļĦæĹ¶ä»£
    0.31
    ultimate
    0.28
    æĹ¶ä»£çļĦ
    0.28
    åĪĨå±Ģ
    0.27
    -era
    0.26
    itel
    0.26
    ellular
    0.26
    æĻĤ代
    0.26
    global
    0.26
    origin
    0.25
    Activations Density 0.016%

    No Known Activations