© 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. 6114
    Prev
    Next
    INDEX
    Explanations

    legal documents

    np_max-act · gemini-2.0-flash

    This neuron fires on formal conclusion or transition words—especially logical connectors like “therefore,” “thus,” “makes,” “conclude,” or other signals of legal‐text logical inference.

    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
    Changing
    -0.07
    ивання
    -0.07
    　 　
    -0.07
    附近
    -0.07
    _Of
    -0.06
    جوی
    -0.06
    .new
    -0.06
     muối
    -0.06
     С
    -0.06
    esan
    -0.06
    POSITIVE LOGITS
    Gl
    0.06
     distilled
    0.06
    .comments
    0.06
    客户
    0.06
    =h
    0.06
     Sour
    0.06
    357
    0.06
    ‌د
    0.06
    xED
    0.06
     ผ
    0.06
    Activations Density 0.012%

    No Known Activations