© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89714
    Prev
    Next
    INDEX
    Explanations

    strong opinions

    np_max-act · gemini-2.0-flash

    This neuron detects emphatic negative evaluative words—strongly opinionated adjectives and nouns (e.g. “audacity,” “unbelievable,” “breathtaking,” “ridiculous”) used to express outrage or criticism.

    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
     tough
    -0.07
     filmpjes
    -0.07
     broth
    -0.07
    .ex
    -0.06
    dea
    -0.06
     verilen
    -0.06
     minced
    -0.06
    Dirty
    -0.06
     glare
    -0.06
     Failed
    -0.06
    POSITIVE LOGITS
    ِب
    0.07
    124
    0.07
    114
    0.06
     leaving
    0.06
     вок
    0.06
     врем
    0.06
    姓名
    0.06
    .Points
    0.06
     governor
    0.06
     disturbed
    0.06
    Activations Density 0.058%

    No Known Activations