© 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. 14-TRANSCODER-HP
    7. 40302
    Prev
    Next
    INDEX
    Explanations

    Article segments

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    HING
    -0.28
    _locals
    -0.27
    EAR
    -0.26
    æİ£
    -0.26
    -web
    -0.26
     web
    -0.25
    Drv
    -0.25
    ä¸įå¾Ĺå·²
    -0.24
     delays
    -0.24
    vak
    -0.24
    POSITIVE LOGITS
    .meta
    0.29
    elpers
    0.26
    è±Ĩçĵ£
    0.26
    åįģ大
    0.25
    æľĪæľ«
    0.25
    acula
    0.25
    åı·
    0.24
    ä»»æĦı
    0.24
    lastName
    0.23
    AuthProvider
    0.23
    Activations Density 0.255%

    No Known Activations