© 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. 39264
    Prev
    Next
    INDEX
    Explanations

    code snippets

    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
    -serif
    -0.29
    ä¹īåĬ¡æķĻèĤ²
    -0.28
    hci
    -0.28
    ]%
    -0.28
     whole
    -0.27
    çľĭä½ł
    -0.26
    whole
    -0.26
    atie
    -0.26
    亦
    -0.25
    ek
    -0.25
    POSITIVE LOGITS
     analyse
    0.26
     veterans
    0.26
     tongues
    0.24
     videog
    0.24
     watchers
    0.24
     academia
    0.24
     labyrinth
    0.24
    Wal
    0.24
     vibes
    0.23
     strife
    0.23
    Activations Density 0.793%

    No Known Activations