© 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. 90069
    Prev
    Next
    INDEX
    Explanations

    random snippets

    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
    Negative Logits
    åĶ³
    -0.28
     objectives
    -0.27
    /sm
    -0.26
    elt
    -0.26
    lander
    -0.25
    olls
    -0.25
    ybrid
    -0.25
     goals
    -0.25
    ä¸Ńåħ³
    -0.24
    els
    -0.24
    POSITIVE LOGITS
    lo
    0.29
    èī¯æĢ§
    0.27
    edList
    0.27
     disproportionately
    0.26
    sort
    0.26
    vendors
    0.25
    ä¸Ģç»ı
    0.25
     **/↵
    0.25
    fect
    0.25
     privately
    0.25
    Activations Density 10.530%

    No Known Activations