© 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. 18-TRANSCODER-HP
    7. 89216
    Prev
    Next
    INDEX
    Explanations

    Medical and programming texts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anking
    -0.29
    quette
    -0.28
    ива
    -0.26
    avra
    -0.26
    éc
    -0.26
    åĨł
    -0.26
     Starter
    -0.25
    aw
    -0.25
    ony
    -0.25
    _________________↵↵
    -0.25
    POSITIVE LOGITS
     also
    0.24
    åIJĥè¿ĩ
    0.24
    ãĥªãĥ³
    0.24
     infused
    0.24
    _Rem
    0.23
     contro
    0.23
    箬
    0.23
    _FT
    0.23
     clue
    0.23
    soever
    0.23
    Activations Density 67.418%

    No Known Activations