© 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. 20-TRANSCODER-HP
    7. 130427
    Prev
    Next
    INDEX
    Explanations

    Technical text/code snippets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    alon
    -0.26
    akening
    -0.25
    entially
    -0.25
    _txn
    -0.25
     Lal
    -0.24
    CharCode
    -0.24
    aid
    -0.24
    rant
    -0.24
    illery
    -0.24
    æĹłå½¢
    -0.23
    POSITIVE LOGITS
    ç±³
    0.28
    Meter
    0.27
    奴éļ¶
    0.26
    -License
    0.26
    强大
    0.25
    丰åİļ
    0.25
    æµģæ·Į
    0.25
    oise
    0.25
    oÅĽÄĩ
    0.25
    agher
    0.24
    Activations Density 1.355%

    No Known Activations