© 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. 19-TRANSCODER-HP
    7. 5284
    Prev
    Next
    INDEX
    Explanations

    peak

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    请è¾ĵåħ¥
    -0.26
     atIndex
    -0.23
    bags
    -0.23
    âĹĦ
    -0.23
    trees
    -0.23
    çīĩ段
    -0.23
    åħħåĪĨ
    -0.22
    .slim
    -0.22
    åºķçĽĺ
    -0.22
    aos
    -0.22
    POSITIVE LOGITS
    æ°Ĺãģ«ãģªãĤĭ
    0.27
    ary
    0.25
     accent
    0.24
    zug
    0.24
     comet
    0.23
    çľĭå¾Ĺ
    0.23
    ä¸įè¿ĩ
    0.23
    ging
    0.23
    ingly
    0.22
    (stderr
    0.22
    Activations Density 0.712%

    No Known Activations