© 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. 18459
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    åĬ²
    -0.32
    aed
    -0.30
    ecx
    -0.27
    åĭģ
    -0.26
    _imm
    -0.25
    *num
    -0.24
    å¿
    -0.24
    å¾Ĺ天
    -0.24
    plätze
    -0.24
    éģĵè·¯
    -0.24
    POSITIVE LOGITS
    iban
    0.30
     deb
    0.29
    ibe
    0.28
    æ´®
    0.26
    antine
    0.26
     survive
    0.25
    ].[
    0.25
    deb
    0.24
    çĶŁåŃĺ
    0.24
    尤为éĩįè¦ģ
    0.24
    Activations Density 1.122%

    No Known Activations

    This feature has no known activations.