© 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 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. 38511
    Prev
    Next
    INDEX
    Explanations

    personal stories

    np_max-act · gemini-2.0-flash

    The neuron responds primarily to first-person references—pronouns like I, my, and possessive ’s indicating the speaker.

    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
     smarter
    -0.07
    �
    -0.07
    -pages
    -0.06
    澳
    -0.06
     smiling
    -0.06
    Video
    -0.06
    Fox
    -0.06
    etail
    -0.06
     कहत
    -0.06
     природ
    -0.06
    POSITIVE LOGITS
    είται
    0.06
    ераль
    0.06
     вт
    0.06
    (PHP
    0.06
     Includes
    0.06
    EDIATE
    0.06
    .getParam
    0.06
    angement
    0.06
    eeee
    0.05
     країни
    0.05
    Activations Density 0.036%

    No Known Activations