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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 4720
    Prev
    Next
    INDEX
    Explanations

    art

    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
    appe
    -0.27
    ᾶ
    -0.25
     arg
    -0.25
    俱
    -0.25
     empty
    -0.25
    æĬµ
    -0.24
    海峡
    -0.24
    ç©¿
    -0.24
    ducer
    -0.24
    html
    -0.23
    POSITIVE LOGITS
    ernes
    0.27
    ::*
    0.26
    çļĦæĪIJåĬŁ
    0.25
    çļĦéĩį大
    0.25
    åłķ
    0.25
    çļĦç§ĺå¯Ĩ
    0.25
    yor
    0.24
    !↵↵↵↵
    0.24
    igans
    0.24
     Marxism
    0.24
    Activations Density 0.011%

    No Known Activations