© 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. 87507
    Prev
    Next
    INDEX
    Explanations

    Computer files and applications

    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
    ķĮ
    -0.27
    æīķãģĦ
    -0.25
     relevance
    -0.24
     noqa
    -0.24
    ,dim
    -0.23
     Scottish
    -0.23
    碰åĪ°
    -0.23
    讣
    -0.23
    åĬŀåħ¬åİħ
    -0.23
    akk
    -0.23
    POSITIVE LOGITS
    骨
    0.26
    ISTER
    0.26
    ä¸īæľŁ
    0.25
    viron
    0.24
     Monsters
    0.24
    ä¾ĭå¤ĸ
    0.24
    Editors
    0.24
    tbl
    0.24
     caster
    0.23
    ird
    0.23
    Activations Density 0.002%

    No Known Activations