© 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. 19-TRANSCODER-HP
    7. 8239
    Prev
    Next
    INDEX
    Explanations

    program

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä»»ä½ķä¸Ģ个
    -0.31
    åħ·
    -0.27
    -any
    -0.26
     hed
    -0.26
    ..."
    -0.24
    anything
    -0.24
     Wired
    -0.24
    ä¸ĭä¾Ĩ
    -0.24
    راج
    -0.24
    éħįèµĦ
    -0.23
    POSITIVE LOGITS
    çĥĽ
    0.27
    ãģĵãĤį
    0.27
    unami
    0.27
    маÑĤ
    0.26
    è§īå¾Ĺèĩªå·±
    0.26
    urt
    0.26
    UMB
    0.25
    eldom
    0.24
    annah
    0.24
     TX
    0.24
    Activations Density 0.010%

    No Known Activations