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. 23-TRANSCODER-HP
    4. 8171
    Prev
    Next
    INDEX
    Explanations

    evaluated

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sets
    -0.30
     
    -0.30
    set
    -0.30
    sets
    -0.30
     impact
    -0.29
     hab
    -0.28
     ne
    -0.28
     ag
    -0.27
     set
    -0.27
    Y
    -0.27
    POSITIVE LOGITS
     ApplicationUser
    0.28
     Returned
    0.27
    æĥ
    0.27
    ici
    0.26
    =color
    0.26
     Equals
    0.26
    games
    0.25
    Equal
    0.25
    [__
    0.25
     inconsist
    0.25
    Activations Density 0.001%

    No Known Activations