© 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. 9-TRANSCODER-HP
    7. 29631
    Prev
    Next
    INDEX
    Explanations

    Punctuation and word fragments

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_9.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.9.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dripping
    -0.30
    {/*
    -0.25
    è¿ĺ好
    -0.25
    è¿ĩæĪ·
    -0.25
    ìĹŃ
    -0.24
     Atatürk
    -0.24
    æµ£
    -0.24
    社ä¼ļèµĦæľ¬
    -0.24
    osos
    -0.23
    樾
    -0.23
    POSITIVE LOGITS
    CellValue
    0.29
    çķĻä¸ĭæĿ¥
    0.28
    reak
    0.28
    eeper
    0.27
     reduced
    0.26
    ningar
    0.25
     Hacker
    0.25
    é¡¿
    0.25
    intér
    0.25
    ularity
    0.25
    Activations Density 0.627%

    No Known Activations