© 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. 112117
    Prev
    Next
    INDEX
    Explanations

    extensive

    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
    inary
    -0.24
    iances
    -0.24
    ispers
    -0.23
    好åıĭ
    -0.23
    stown
    -0.23
    dish
    -0.22
     Currently
    -0.22
     Convenience
    -0.22
    eselect
    -0.22
    FI
    -0.22
    POSITIVE LOGITS
     calendar
    0.27
    combe
    0.26
     enough
    0.26
    è¦ĥ
    0.26
     role
    0.26
    å§Ķåijĺä¼ļ
    0.26
     (>
    0.25
    orre
    0.25
    endor
    0.25
    HEMA
    0.24
    Activations Density 0.232%

    No Known Activations