© 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. 19188
    Prev
    Next
    INDEX
    Explanations

    solving problems

    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
    èĪĶ
    -0.25
    沫
    -0.23
    udit
    -0.23
    å·¢
    -0.23
    æĬĬæīĭ
    -0.23
     eat
    -0.23
     undert
    -0.23
    ycles
    -0.22
    éĩįå¿ĥ
    -0.22
    ãĤ¢ãĤ¯
    -0.22
    POSITIVE LOGITS
    解
    1.16
     sol
    1.03
     Sol
    0.93
    sol
    0.87
     solving
    0.84
    解åĨ³
    0.84
     solve
    0.82
     解
    0.80
     SOL
    0.80
    _SOL
    0.79
    Activations Density 13.369%

    No Known Activations