© 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. 87451
    Prev
    Next
    INDEX
    Explanations

    Question words

    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
    haft
    -0.26
    annes
    -0.26
    åŃĶ
    -0.26
    è¿Ļä¸Ģ
    -0.25
    eks
    -0.25
     scratch
    -0.25
    æ³ķåĪ¶
    -0.25
    çľī
    -0.25
    æīĢ说
    -0.24
    олоÑģ
    -0.24
    POSITIVE LOGITS
     did
    0.38
     does
    0.38
     do
    0.38
    did
    0.33
    éļ¾éģĵ
    0.31
     dose
    0.30
    do
    0.30
    rad
    0.28
     doe
    0.28
    does
    0.27
    Activations Density 0.163%

    No Known Activations