© 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. 27-TRANSCODER-HP
    7. 9834
    Prev
    Next
    INDEX
    Explanations

    explanation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ä¸ĸ家
    -0.24
    nore
    -0.22
    (_:
    -0.22
     reelection
    -0.22
    çī¹æ´¾åijĺ
    -0.22
    picker
    -0.22
    BASH
    -0.22
    梯éĺŁ
    -0.22
    è°£
    -0.22
    UMENT
    -0.22
    POSITIVE LOGITS
     provided
    0.30
    æīĢæıIJä¾Ľ
    0.27
    train
    0.25
    OTS
    0.25
    ful
    0.24
    ble
    0.24
     crisp
    0.24
    å±Ģ
    0.23
    nts
    0.23
    bin
    0.23
    Activations Density 0.045%

    No Known Activations