© 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. 25-TRANSCODER-HP
    7. 108878
    Prev
    Next
    INDEX
    Explanations

    Code and related documentation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ega
    -0.33
    omm
    -0.28
    erg
    -0.28
    sgi
    -0.28
    åıĥ
    -0.27
    efa
    -0.27
    åıĤ
    -0.26
    ÃŃt
    -0.26
     ton
    -0.26
    og
    -0.26
    POSITIVE LOGITS
    linky
    0.31
    ainter
    0.28
    adesh
    0.27
    uai
    0.26
    -boy
    0.26
    .InputStreamReader
    0.26
    HASH
    0.26
    -bel
    0.26
    æ³ł
    0.24
    ledged
    0.24
    Activations Density 0.951%

    No Known Activations