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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 98294
    Prev
    Next
    INDEX
    Explanations

    main

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĶ²
    -0.29
    个å¤ļ
    -0.28
    lish
    -0.28
    Gui
    -0.26
    +',
    -0.25
    tur
    -0.25
    phin
    -0.25
    ushing
    -0.24
    felt
    -0.23
    .touch
    -0.23
    POSITIVE LOGITS
    主æīĵ
    0.28
    mı
    0.26
     principle
    0.26
    principal
    0.26
    åİŁåĪĻ
    0.25
    ä¸ĢåľĪ
    0.25
    soever
    0.25
    enerator
    0.25
    ICC
    0.25
    PropertyParams
    0.24
    Activations Density 0.026%

    No Known Activations