© 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. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 127050
    Prev
    Next
    INDEX
    Explanations

    legal documents

    np_max-act · gemini-2.0-flash

    This neuron is recognizing tokens based on their position in the document, activating most strongly on words in the middle sections of these legal texts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    care
    -0.07
     мыш
    -0.07
    .basic
    -0.06
     cigars
    -0.06
     Slovenia
    -0.06
     угод
    -0.06
    Tek
    -0.06
    .ON
    -0.06
    函数
    -0.06
     wheels
    -0.06
    POSITIVE LOGITS
     Conan
    0.06
    .preview
    0.06
    碎
    0.06
     TASK
    0.06
    [root
    0.06
     jLabel
    0.06
     payouts
    0.06
     ráno
    0.06
    HAL
    0.06
     Crushing
    0.06
    Activations Density 0.004%

    No Known Activations