© 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. 88317
    Prev
    Next
    INDEX
    Explanations

    steam/stein

    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
    ersions
    -0.31
    hang
    -0.25
     Obtain
    -0.24
    åĥįãģį
    -0.24
     chambers
    -0.24
    sic
    -0.24
    é¢ŀ
    -0.23
    Uint
    -0.23
     yap
    -0.23
    isons
    -0.23
    POSITIVE LOGITS
     Arabia
    0.30
     Arabian
    0.26
    éķIJ
    0.25
    ive
    0.24
    kin
    0.24
    èı°
    0.23
    grav
    0.23
    ãģ§ãģį
    0.23
    èı²å°Ķ
    0.23
    亳
    0.22
    Activations Density 0.197%

    No Known Activations