© 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. 12-TRANSCODER-HP
    7. 23592
    Prev
    Next
    INDEX
    Explanations

    Origin, source, extraction (from)

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     start
    -0.34
    èµ·çĤ¹
    -0.31
    ä¸Ģå¼Ģå§ĭ
    -0.31
    ä¸Ģå¼Ģå§ĭå°±
    -0.30
     starting
    -0.30
    å¼Ģå§ĭ
    -0.30
     starts
    -0.29
     started
    -0.29
     beginnings
    -0.29
     begin
    -0.29
    POSITIVE LOGITS
     stead
    0.28
    Tro
    0.28
    åĨł
    0.27
    kins
    0.27
    setData
    0.26
    çĬĩ
    0.26
    è¿Ļä¹Īåģļ
    0.25
    panse
    0.25
    Cpp
    0.24
    Higher
    0.24
    Activations Density 0.003%

    No Known Activations