© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 162873
    Prev
    Next
    INDEX
    Explanations

    Kong, Tong

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -note
    -0.27
    atos
    -0.26
    èľĤ
    -0.24
    ħ§
    -0.24
    inos
    -0.23
     taraf
    -0.23
    LY
    -0.23
    篷
    -0.23
    ½æķ°
    -0.23
    Notes
    -0.22
    POSITIVE LOGITS
    e
    0.32
    ç¢ĩ
    0.31
    cil
    0.29
    pile
    0.29
    ea
    0.28
    eon
    0.28
    eur
    0.27
    ection
    0.27
    ein
    0.27
    ei
    0.26
    Activations Density 0.209%

    No Known Activations