© 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. 19-TRANSCODER-HP
    7. 9116
    Prev
    Next
    INDEX
    Explanations

    itone

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Islands
    -0.29
    tag
    -0.25
    beck
    -0.25
    åħ³éĶ®æŶåĪ»
    -0.24
     islands
    -0.24
    à®ľ
    -0.24
    èĿ¶
    -0.23
    æľ¬èĥ½
    -0.23
    èŀįåħ¥
    -0.23
    ritic
    -0.22
    POSITIVE LOGITS
    åĬŀ
    0.29
    chr
    0.26
    /animate
    0.26
    pus
    0.25
    chs
    0.24
     Weiner
    0.24
    ç»ĨåĪĨ
    0.23
    ç½ij讯
    0.23
    wan
    0.23
     corpus
    0.23
    Activations Density 0.153%

    No Known Activations