© 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. 142011
    Prev
    Next
    INDEX
    Explanations

    Stack Exchange questions/edits

    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
    å¹³è¡Į
    -0.27
    çĶŁåij½çļĦ
    -0.27
     оÑĤз
    -0.24
    æ¿¡
    -0.24
     outs
    -0.24
     yearly
    -0.24
    æ¯ıå¹´éĥ½
    -0.23
    Charsets
    -0.23
    åħ¨èĩªåĬ¨
    -0.23
    ä¸ĢçĶŁ
    -0.23
    POSITIVE LOGITS
    æĶ¹ä¸º
    0.52
    ä¿®æĶ¹
    0.49
    æĶ¹
    0.47
     revised
    0.46
    è°ĥæķ´
    0.44
    修订
    0.43
    åİŁæĿ¥çļĦ
    0.43
     replaced
    0.43
     replacing
    0.42
    æĸ°çļĦ
    0.42
    Activations Density 63.687%

    No Known Activations