© 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. 35-TRANSCODER-HP
    7. 40007
    Prev
    Next
    INDEX
    Explanations

    degree symbol

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     affid
    -0.26
    âĢ¦”↵↵
    -0.26
    "↵↵↵
    -0.25
     conspir
    -0.24
    .Migrations
    -0.24
    ]]↵↵
    -0.24
     burge
    -0.24
     brawl
    -0.23
     inform
    -0.23
    ."));↵
    -0.23
    POSITIVE LOGITS
    Fo
    0.41
    Fx
    0.39
    Fc
    0.38
    Fs
    0.36
    Ce
    0.34
    Cb
    0.34
    Cx
    0.34
    Ц
    0.34
    Xi
    0.34
    Fa
    0.33
    Activations Density 0.167%

    No Known Activations