© 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. 97393
    Prev
    Next
    INDEX
    Explanations

    Irish

    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
    eper
    -0.30
    eer
    -0.27
    ehr
    -0.27
    å¸Ń
    -0.26
     others
    -0.25
    e
    -0.25
    stance
    -0.24
    åįıä¼ļä¼ļåijĺ
    -0.24
    gets
    -0.24
    hey
    -0.24
    POSITIVE LOGITS
    ment
    0.30
    ellaneous
    0.29
    theless
    0.27
    ibe
    0.27
    ness
    0.27
    andom
    0.27
    åĮĸ
    0.27
    sonian
    0.26
    ingly
    0.26
    onical
    0.26
    Activations Density 0.061%

    No Known Activations