© 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. 109855
    Prev
    Next
    INDEX
    Explanations

    then

    np_max-act · gemini-2.0-flash

    This neuron fires on the special protocol/control tokens (e.g. the `<|…|>` markers and other header/separator tokens) that delineate the prompt structure.

    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
    Explore
    -0.06
    iris
    -0.06
    (tableView
    -0.06
     Bradley
    -0.06
     believable
    -0.06
     놓
    -0.06
    امین
    -0.06
    ursions
    -0.06
    .DataGridViewTextBoxColumn
    -0.06
    forums
    -0.06
    POSITIVE LOGITS
     Emb
    0.07
     IBM
    0.07
     velit
    0.06
     cn
    0.06
     specifics
    0.06
     OSS
    0.06
    ocities
    0.06
     activate
    0.06
    ็นผ
    0.06
     hurry
    0.06
    Activations Density 0.003%

    No Known Activations