© 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. 88617
    Prev
    Next
    INDEX
    Explanations

    distinct

    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
    åĬłä»¥
    -0.28
    åİ»åģļ
    -0.28
     purposes
    -0.27
    à¹Ħà¸Ľ
    -0.26
    ).(
    -0.26
    æ¿ij
    -0.25
     directive
    -0.25
     kolej
    -0.25
    ÅĽl
    -0.24
     cá»ij
    -0.24
    POSITIVE LOGITS
    è°ģéĥ½
    0.30
    orama
    0.29
    éĤ£æ®µ
    0.28
    uzzi
    0.26
    iris
    0.26
    <V
    0.25
    èĦĪ
    0.25
    anje
    0.24
    anja
    0.24
    amient
    0.24
    Activations Density 0.014%

    No Known Activations