© 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. 18886
    Prev
    Next
    INDEX
    Explanations

    Code and computing

    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
     hash
    -0.66
     hashes
    -0.61
    hash
    -0.60
    _hash
    -0.57
    	hash
    -0.57
    Hash
    -0.56
     HASH
    -0.54
    _hashes
    -0.54
    (hash
    -0.54
     Hash
    -0.54
    POSITIVE LOGITS
    styled
    0.33
    eware
    0.31
    themes
    0.25
    angu
    0.25
    oji
    0.25
    æ±ŁéŨ
    0.25
    atism
    0.25
    å®ĺ
    0.24
    UGHT
    0.24
    _DISCONNECT
    0.24
    Activations Density 1.226%

    No Known Activations