© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 163405
    Prev
    Next
    INDEX
    Explanations

    Multiple/Uncertain reasons or amount

    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
    好人
    -0.27
     EACH
    -0.26
    æĮĩæľĽ
    -0.26
    ä¸Ģç¢Ĺ
    -0.26
    åĨ¢
    -0.25
    pons
    -0.25
    eko
    -0.25
     @[
    -0.25
    ookie
    -0.25
     wid
    -0.25
    POSITIVE LOGITS
     manner
    0.27
    ãĤĦãĤĬ
    0.26
    ettes
    0.25
    erÃŃa
    0.24
     fashion
    0.24
    illos
    0.24
     relat
    0.24
    (sf
    0.24
    itters
    0.23
    ë±ħ
    0.23
    Activations Density 0.004%

    No Known Activations