© 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
    2. Qwen3-4B
    3. 22-TRANSCODER-HP
    4. 97800
    Prev
    Next
    INDEX
    Explanations

    Descriptions and specifications

    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
    cape
    -0.26
    RIX
    -0.26
    бÑĥÑĢ
    -0.25
    undra
    -0.25
    ĵį
    -0.25
    裴
    -0.24
    GET
    -0.24
     Ste
    -0.24
    å®ĥçļĦ
    -0.24
    åĪĨè§£
    -0.24
    POSITIVE LOGITS
     crossed
    0.30
    озв
    0.28
    OfSize
    0.26
    æĮ¯
    0.25
    ä¼ļå¼ķèµ·
    0.25
    minent
    0.24
    xad
    0.24
    æīĢä»¥åľ¨
    0.24
    iker
    0.24
    leton
    0.24
    Activations Density 0.001%

    No Known Activations