Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 100211
    Prev
    Next
    INDEX
    Explanations

    <

    np_max-act · gemini-2.0-flash

    The neuron fires on HTML markup syntax—that is, tokens belonging to tags and attributes (e.g. the “<…>” brackets and tag names).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    PHP code for creating a login system with user authentication.

    oai_token-act-pair · gpt-4o-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
    ="<?=$
    -0.07
    eus
    -0.06
     };↵↵↵
    -0.06
    ]=(
    -0.06
    OUTPUT
    -0.06
    EXAMPLE
    -0.06
    cdot
    -0.06
    espoň
    -0.06
    Ke
    -0.06
     pavement
    -0.06
    POSITIVE LOGITS
     واج
    0.07
    (off
    0.07
    ़ा
    0.06
     싶
    0.06
    ắ
    0.06
     terug
    0.06
    .out
    0.06
    iế
    0.06
    ौं
    0.06
    玩
    0.06
    Activations Density 0.018%

    No Known Activations