© 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. 88256
    Prev
    Next
    INDEX
    Explanations

    Technical descriptions and references

    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
     Nagar
    -0.25
    æ³¥åľŁ
    -0.25
    æī¶æĮģ
    -0.23
     ÑħоÑĤел
    -0.23
    éĥ½åºĶ该
    -0.23
    âĢ¦↵↵↵
    -0.23
    agram
    -0.23
    eway
    -0.23
     nag
    -0.22
     Dur
    -0.22
    POSITIVE LOGITS
    æĪij们çļĦ
    0.28
    ç®Ģ
    0.28
     Alv
    0.26
     ours
    0.26
    ç°¡
    0.26
     syn
    0.25
    .ErrorCode
    0.24
    ALA
    0.24
    é¦Ķ
    0.24
    éªĦ
    0.24
    Activations Density 0.008%

    No Known Activations