© 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. 6136
    Prev
    Next
    INDEX
    Explanations

    Code/website headers

    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
    orted
    -0.28
    zier
    -0.26
    elim
    -0.26
    éĹ®ä»ĸ
    -0.25
    anding
    -0.24
    Pass
    -0.24
    nim
    -0.24
    ner
    -0.24
    irc
    -0.24
    orts
    -0.24
    POSITIVE LOGITS
    ç²ī
    0.25
    æĺ¨
    0.25
    :uint
    0.25
    å¥ķ
    0.25
    usher
    0.25
    婳
    0.24
    å¸ĥ
    0.24
     Step
    0.23
     iota
    0.23
    éķ¶
    0.23
    Activations Density 0.026%

    No Known Activations