© 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. 6134
    Prev
    Next
    INDEX
    Explanations

    carcinogen

    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
     allowance
    -0.27
    ed
    -0.27
    æŃ¤åŁºç¡Ģä¸Ĭ
    -0.26
    usal
    -0.25
    neath
    -0.24
    erson
    -0.23
    çľĭå¾ħ
    -0.23
    routine
    -0.23
    fitness
    -0.23
    mia
    -0.23
    POSITIVE LOGITS
     abroad
    0.25
    çĸ½
    0.24
    wat
    0.23
     Weeks
    0.23
    offs
    0.23
    rons
    0.23
    oons
    0.23
    çIJħ
    0.23
    itory
    0.22
    å¾Ĵ
    0.22
    Activations Density 1.196%

    No Known Activations