© 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. 19-TRANSCODER-HP
    7. 6318
    Prev
    Next
    INDEX
    Explanations

    novelty or uniqueness

    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
    -errors
    -0.29
    embro
    -0.26
    .photos
    -0.26
    被åĪ¤
    -0.26
     lambda
    -0.25
    زاÙĦ
    -0.25
    æŨ
    -0.25
    confirmation
    -0.25
    IMPORT
    -0.24
     gains
    -0.24
    POSITIVE LOGITS
     rare
    0.40
     unique
    0.39
    åīįæīĢæľªæľī
    0.39
    没è§ģè¿ĩ
    0.36
    åīįæīĢæľª
    0.36
    unique
    0.35
     unprecedented
    0.35
    åıªæľī
    0.35
    ç½ķè§ģ
    0.35
    没æľī人
    0.34
    Activations Density 0.300%

    No Known Activations