© 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. 18-TRANSCODER-HP
    7. 89964
    Prev
    Next
    INDEX
    Explanations

    I am/have

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æŃ£éĿ¢
    -0.29
    èģĶ
    -0.25
    isl
    -0.25
    notated
    -0.25
    ()='
    -0.24
    .jd
    -0.24
    NavController
    -0.23
     ÑĥÑĩеб
    -0.23
    ä¹ĭ家
    -0.23
    uchen
    -0.23
    POSITIVE LOGITS
    Thr
    0.26
    ,dim
    0.24
    .TIM
    0.24
    çĽ¸ä¿¡å¤§å®¶
    0.24
     gsi
    0.24
    Limits
    0.23
     Limits
    0.23
    inant
    0.23
    Growing
    0.23
     Gian
    0.23
    Activations Density 48.180%

    No Known Activations