© 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. 30815
    Prev
    Next
    INDEX
    Explanations

    would

    np_max-act · gemini-2.0-flash

    The neuron responds to occurrences of second-person pronouns and associated auxiliary/modal expressions (e.g. "you," "would," "be") used in direct address.

    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
    /art
    -0.07
     GlobalKey
    -0.06
     Rew
    -0.06
     domicile
    -0.06
    Embedded
    -0.06
     rubber
    -0.06
    /server
    -0.06
    -0.06
    ères
    -0.06
    .Currency
    -0.06
    POSITIVE LOGITS
     would
    0.07
    鳴
    0.07
    unca
    0.07
     všichni
    0.06
     всі
    0.06
     misery
    0.06
     lety
    0.06
     پیشنه
    0.06
    ่าเป
    0.06
     consistent
    0.06
    Activations Density 0.038%

    No Known Activations