Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Qwen3-4B
    3. 11-TRANSCODER-HP
    4. 4000
    Prev
    Next
    INDEX
    Explanations

    varied content sources

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_11.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.11.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     exercise
    -0.26
    oger
    -0.25
    å£ģ
    -0.25
    æĥ¨
    -0.25
     joint
    -0.25
    强åζ
    -0.25
     Fan
    -0.24
    pheric
    -0.24
     wys
    -0.24
    _go
    -0.24
    POSITIVE LOGITS
    reated
    0.30
    çļĦè§Ĵèī²
    0.27
    	reply
    0.27
    weigh
    0.27
    rough
    0.27
    rende
    0.26
    rend
    0.26
    (reply
    0.25
    é«Ń
    0.24
    aju
    0.24
    Activations Density 0.295%

    No Known Activations