© 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. 31-TRANSCODER-HP
    7. 18345
    Prev
    Next
    INDEX
    Explanations

    conj

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĢ»è¾ij
    -0.36
    Logic
    -0.31
    false
    -0.31
    logic
    -0.30
     logical
    -0.29
    羣åģĩ
    -0.29
    agger
    -0.28
    Logical
    -0.28
    False
    -0.27
    æīij
    -0.27
    POSITIVE LOGITS
     conj
    0.35
    conj
    0.33
    HO
    0.28
     Conj
    0.28
    .cn
    0.28
    hoc
    0.28
    CN
    0.27
     KN
    0.27
    rollo
    0.26
    uren
    0.26
    Activations Density 4.436%

    No Known Activations