© 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. 88822
    Prev
    Next
    INDEX
    Explanations

    names and parts of names

    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
    -esque
    -0.33
    them
    -0.30
    crest
    -0.29
    istic
    -0.28
    testing
    -0.28
    istically
    -0.27
    zd
    -0.27
    sville
    -0.27
    aceous
    -0.26
    ноÑģÑĤей
    -0.26
    POSITIVE LOGITS
    (es
    0.35
    es
    0.32
    ive
    0.29
    ×
    0.29
    ed
    0.27
    udio
    0.26
    imoto
    0.25
    ingly
    0.24
    band
    0.24
    esor
    0.24
    Activations Density 0.449%

    No Known Activations