© 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. 32-TRANSCODER-HP
    7. 42912
    Prev
    Next
    INDEX
    Explanations

    Historic places

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iques
    -0.27
    梯
    -0.27
    iores
    -0.26
    .cg
    -0.26
    gradient
    -0.25
    人æ°ijéĵ¶è¡Į
    -0.25
    cé
    -0.25
    rice
    -0.25
    _bases
    -0.24
    çĶ³è¯·äºº
    -0.24
    POSITIVE LOGITS
     Hot
    0.30
    udio
    0.27
    cold
    0.27
     polys
    0.27
     Saf
    0.27
    Hot
    0.27
    loff
    0.25
    íķ«
    0.25
     Minds
    0.25
    ature
    0.25
    Activations Density 0.004%

    No Known Activations