© 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
    2. Qwen3-4B
    3. 33-TRANSCODER-HP
    4. 53695
    Prev
    Next
    INDEX
    Explanations

    Code snippets and random text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     and
    -0.53
     so
    -0.42
    ,
    -0.40
     ;
    -0.40
    ,”
    -0.40
    ;
    -0.38
     but
    -0.37
    ;.
    -0.34
     as
    -0.33
    .”
    -0.33
    POSITIVE LOGITS
    â̦)↵↵
    0.26
    !).↵↵
    0.25
     AUTHORS
    0.25
    éĺ¿å§¨
    0.24
    Means
    0.24
    åķĨä¼ļ
    0.24
    IRA
    0.24
     mỹ
    0.23
    amazon
    0.23
    ä¹ħäºĨ
    0.23
    Activations Density 0.032%

    No Known Activations