© 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. 42770
    Prev
    Next
    INDEX
    Explanations

    Quotation marks

    np_max-act · gemini-2.0-flash

    The neuron specializes in detecting XML/HTML prolog or header declarations (e.g. the `<?xml …?>`, DOCTYPE, or content‐type meta tags).

    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
     wisely
    -0.06
     resumes
    -0.06
     Flu
    -0.06
    elt
    -0.06
    arr
    -0.06
    13
    -0.06
    anto
    -0.06
    Docs
    -0.06
    14
    -0.06
     dys
    -0.06
    POSITIVE LOGITS
    、《
    0.06
     діяльності
    0.06
     ''↵
    0.06
     mnemonic
    0.06
    tel
    0.06
    گران
    0.06
    成为
    0.06
    .WriteAll
    0.06
     줄
    0.06
     ничего
    0.06
    Activations Density 0.001%

    No Known Activations