© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 96464
    Prev
    Next
    INDEX
    Explanations

    legal/historical/political contexts

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å¾Ĥ
    -0.26
    arks
    -0.25
    AGO
    -0.25
    è£Ł
    -0.25
    enet
    -0.25
     Bedford
    -0.24
    Ã¼ÅŁ
    -0.24
    lemn
    -0.23
     Todd
    -0.23
    PTS
    -0.23
    POSITIVE LOGITS
    æ¯Ĥ
    0.26
     ApiException
    0.26
    ½Ķ
    0.26
     Conference
    0.25
     dur
    0.25
     Gap
    0.25
    Conference
    0.25
     conference
    0.24
    人åıĤ
    0.24
    rie
    0.24
    Activations Density 0.900%

    No Known Activations