© 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. 38448
    Prev
    Next
    INDEX
    Explanations

    experiments and procedures

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of experimental controls or reference standards (e.g. “served as a control,” “used as the comparison standard,” “included as a reference”).

    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
     worn
    -0.07
    .Ο
    -0.07
    _sibling
    -0.06
    contracts
    -0.06
     wife
    -0.06
    Checked
    -0.06
    呼
    -0.06
     boots
    -0.06
     줄
    -0.06
    .dex
    -0.06
    POSITIVE LOGITS
    _IB
    0.06
    .Ab
    0.06
    _V
    0.06
    formik
    0.06
     生
    0.06
    (dirname
    0.06
     denomination
    0.06
     RTWF
    0.06
     ↵↵↵↵↵
    0.06
     массив
    0.06
    Activations Density 0.074%

    No Known Activations