© 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. 31-TRANSCODER-HP
    7. 19161
    Prev
    Next
    INDEX
    Explanations

    c

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Retention
    -0.30
    ewe
    -0.28
    æĿ¥çľĭ
    -0.26
    eh
    -0.25
    eten
    -0.25
    RET
    -0.25
     retention
    -0.25
    Ṫ
    -0.25
    VIEW
    -0.24
    .onView
    -0.24
    POSITIVE LOGITS
    cr
    0.30
    æ°´æĻ¶
    0.29
    æĭĦ
    0.28
    -cr
    0.27
    cry
    0.27
     cloud
    0.26
    ä¸ĭæ²ī
    0.26
    cloud
    0.25
    cran
    0.25
    Ļ¶
    0.25
    Activations Density 0.162%

    No Known Activations