© 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. 33-TRANSCODER-HP
    7. 53601
    Prev
    Next
    INDEX
    Explanations

    foreign languages

    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
     them
    -0.28
    èѬ
    -0.27
    .er
    -0.27
    awe
    -0.27
     wor
    -0.26
    稳å®ļæĢ§
    -0.26
    uo
    -0.25
    yer
    -0.25
    æľªç»ı
    -0.25
    å¦Ĥå®ŀ
    -0.25
    POSITIVE LOGITS
    they
    0.37
     die
    0.36
     They
    0.36
     THEY
    0.34
     они
    0.34
     they
    0.33
    They
    0.33
     das
    0.31
     den
    0.30
    aches
    0.29
    Activations Density 0.019%

    No Known Activations