© 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. 23584
    Prev
    Next
    INDEX
    Explanations

    marrow

    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
    ports
    -0.29
     Await
    -0.26
    Grow
    -0.26
    åīĬ
    -0.25
    åıijå±ķçIJĨ念
    -0.24
     Translate
    -0.24
     Ses
    -0.24
     Lion
    -0.24
    GO
    -0.24
    amines
    -0.24
    POSITIVE LOGITS
    veau
    0.29
     cheek
    0.27
    ãĤ£
    0.26
     Starr
    0.26
    reveal
    0.26
    .pat
    0.25
    reative
    0.25
    èĩ´æķ¬
    0.25
    orro
    0.25
    cloak
    0.24
    Activations Density 0.004%

    No Known Activations