© 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. 20-TRANSCODER-HP
    7. 62838
    Prev
    Next
    INDEX
    Explanations

    environmental adaptation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    çĵ¶
    -0.28
    vertex
    -0.26
    åľ°æ¯¯
    -0.26
    arrant
    -0.25
     Neck
    -0.24
    ÑģÑĤвеннÑĭй
    -0.24
    æĺ¯éŁ©åĽ½
    -0.24
    æ¡Ī
    -0.24
    çijĽ
    -0.24
    iliation
    -0.23
    POSITIVE LOGITS
     omin
    0.31
     easiest
    0.31
     underage
    0.28
    æľĢ容æĺĵ
    0.28
     OMIT
    0.26
    è¶ģ
    0.26
    .diff
    0.25
     indiv
    0.25
    berra
    0.25
    .$.
    0.25
    Activations Density 13.044%

    No Known Activations