© 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. 19-TRANSCODER-HP
    7. 7704
    Prev
    Next
    INDEX
    Explanations

    Code and programming

    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
    è¢ĭ
    -0.29
    ays
    -0.28
    è¿Ļä¹Īå¤ļ
    -0.27
    ¹Ħ
    -0.27
    ijke
    -0.26
    bach
    -0.26
    \:
    -0.25
    海滨
    -0.25
    /Core
    -0.24
    backs
    -0.24
    POSITIVE LOGITS
     rever
    0.30
    acin
    0.27
    Ñĭваем
    0.26
    ration
    0.25
     mentally
    0.25
    ãĤ®ãĥ£
    0.24
    uffer
    0.24
    çĥ©
    0.24
    ç½ij
    0.24
    lication
    0.23
    Activations Density 0.793%

    No Known Activations