© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 6525
    Prev
    Next
    INDEX
    Explanations

    programming debugging

    np_max-act · gemini-2.0-flash

    The neuron fires on qualifying or emphasizing adjectives and adverbs (e.g. “only,” “specific,” “certain,” “odd,” “particular”).

    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
    cpy
    -0.07
     vbox
    -0.07
    ินเด
    -0.06
    (reordered
    -0.06
    화를
    -0.06
     Rhodes
    -0.06
    321
    -0.06
     contest
    -0.06
     Initi
    -0.06
    _boxes
    -0.06
    POSITIVE LOGITS
     ř
    0.07
    @class
    0.07
     #@
    0.07
    testdata
    0.06
    ÷
    0.06
    predictions
    0.06
     порт
    0.06
     pagina
    0.06
     revelation
    0.06
    ])/
    0.06
    Activations Density 0.129%

    No Known Activations