© 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. 130368
    Prev
    Next
    INDEX
    Explanations

    Separating

    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
    ÑĥÑĪки
    -0.28
    çĬ¹
    -0.28
    çĮ¶
    -0.26
    Å«
    -0.26
    take
    -0.25
    éĤ£æł·
    -0.24
     take
    -0.24
     tak
    -0.24
    ited
    -0.24
    amber
    -0.24
    POSITIVE LOGITS
     wah
    0.27
    acea
    0.26
    arf
    0.25
    nero
    0.25
    infinity
    0.25
    ecast
    0.25
    cas
    0.25
     geme
    0.24
    rix
    0.24
    笳
    0.23
    Activations Density 0.012%

    No Known Activations