© 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. 97842
    Prev
    Next
    INDEX
    Explanations

    numbers and abbreviations

    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
     Somalia
    -0.30
     Indonesian
    -0.30
    éĿŀæ´²
    -0.30
    å½ĵåľ°äºº
    -0.30
     Kok
    -0.30
     regional
    -0.29
     Somali
    -0.29
     Nigerian
    -0.29
    æľ¬åĽ½
    -0.29
    Africa
    -0.28
    POSITIVE LOGITS
    swick
    0.27
    elsey
    0.26
    .mit
    0.26
    æīĴ
    0.26
    orns
    0.26
    /javascript
    0.25
     assignable
    0.25
     Slee
    0.25
    abler
    0.24
     Neal
    0.24
    Activations Density 0.010%

    No Known Activations