© 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. 20-TRANSCODER-HP
    7. 130756
    Prev
    Next
    INDEX
    Explanations

    questions and evaluations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çªģåĩºéĹ®é¢ĺ
    -0.33
    åĢ¼å¾Ĺåħ³æ³¨
    -0.30
    éĩįçĤ¹
    -0.28
    æĦŁåħ´è¶£
    -0.27
    æĦŁåħ´è¶£çļĦ
    -0.27
    æĭ©
    -0.26
    çĦ¦çĤ¹
    -0.26
     Lent
    -0.25
    ãĥĿãĤ¤ãĥ³ãĥĪ
    -0.25
    çľļ
    -0.25
    POSITIVE LOGITS
    fully
    0.28
    loid
    0.27
    ä¸įçĦ¶
    0.27
    åIJ¬äºĨ
    0.26
    æĪĺäºī
    0.24
    æį¢äºĨ
    0.24
    çģ«çĪĨ
    0.24
    egan
    0.24
    Append
    0.23
    oders
    0.23
    Activations Density 36.928%

    No Known Activations