© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 113063
    Prev
    Next
    INDEX
    Explanations

    am

    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
    Negative Logits
    Ñģл
    -0.28
    .dsl
    -0.25
    æݬ
    -0.25
    åľ¨ä¹İ
    -0.24
    proper
    -0.23
     Publi
    -0.23
    Civil
    -0.23
    aN
    -0.23
    egin
    -0.23
    éĹªåħī
    -0.23
    POSITIVE LOGITS
    unch
    0.32
    anh
    0.27
    éĴĿ
    0.27
     ways
    0.27
    ways
    0.27
    ointments
    0.26
    way
    0.26
    çĴĭ
    0.25
    ninger
    0.25
    ulate
    0.25
    Activations Density 0.028%

    No Known Activations