© 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. 111752
    Prev
    Next
    INDEX
    Explanations

    commas and and

    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
     toch
    -0.27
    åİŁæłĩé¢ĺ
    -0.26
     sweating
    -0.26
    utsch
    -0.26
    åĩºæ±Ĺ
    -0.25
    chs
    -0.25
     colspan
    -0.25
    ç¡ķ
    -0.25
    åĪĹ
    -0.25
    неÑĨ
    -0.24
    POSITIVE LOGITS
     cleans
    0.27
    .cleanup
    0.26
     administrations
    0.26
    isse
    0.25
    å½ĵæĹ¶çļĦ
    0.25
    Minus
    0.25
    说æĺ¯
    0.24
     #####
    0.24
     Clean
    0.24
     article
    0.24
    Activations Density 0.046%

    No Known Activations