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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 6510
    Prev
    Next
    INDEX
    Explanations

    pairing/combining things

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸įæŃ¢
    -0.27
    åĽ´ç»ķ
    -0.25
    TEE
    -0.25
    tec
    -0.24
     vary
    -0.24
    -stop
    -0.24
    &oacute
    -0.24
    Theta
    -0.23
    åħ¨éĥ¨
    -0.23
    èĬ¸
    -0.23
    POSITIVE LOGITS
     pairing
    0.50
     pairs
    0.43
    Pairs
    0.41
    æIJŃéħį
    0.39
    _pairs
    0.38
    pairs
    0.34
     paired
    0.33
     friendship
    0.32
     Pair
    0.32
     pairwise
    0.32
    Activations Density 0.050%

    No Known Activations