© 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. 74774
    Prev
    Next
    INDEX
    Explanations

    code and queries

    np_max-act · gemini-2.0-flash

    The neuron detects instructional or prompt‐related terms in a numerical QA task setup—words like “arithmetic,” “program,” “question,” “Answer,” and “Generate” that form the meta‐instructions for composing contexts and calculating answers.

    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
    .which
    -0.07
    .findall
    -0.07
    lightbox
    -0.06
    /compiler
    -0.06
    _display
    -0.06
     Returned
    -0.06
    輸
    -0.06
    woff
    -0.06
     report
    -0.06
    	find
    -0.06
    POSITIVE LOGITS
     κορ
    0.07
    (ierr
    0.06
     denote
    0.06
    λιά
    0.06
    _FORWARD
    0.06
    ROTO
    0.06
    =sum
    0.06
     largo
    0.06
    errs
    0.06
    /usr
    0.06
    Activations Density 0.008%

    No Known Activations