© 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. 4871
    Prev
    Next
    INDEX
    Explanations

    Architecture and technology texts

    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
    IFO
    -0.29
    RowIndex
    -0.26
    inux
    -0.26
    dac
    -0.24
    istring
    -0.24
    -dem
    -0.24
    rift
    -0.24
    åŃ£åIJİ
    -0.23
    activ
    -0.23
    paced
    -0.23
    POSITIVE LOGITS
    åįļ
    0.25
     æŀ
    0.25
    redi
    0.24
     "
    0.24
     perfect
    0.24
     bian
    0.23
    åĽŃ
    0.23
    æıı
    0.23
    ä½Ĩæĺ¯ä»ĸ
    0.23
    ä¹ĭ
    0.23
    Activations Density 0.045%

    No Known Activations