© 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
    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. 6914
    Prev
    Next
    INDEX
    Explanations

    first-person pronoun

    np_max-act · gemini-2.0-flash

    This neuron flags tokens that occur in the model’s own (assistant) output text (i.e. it activates on tokens in assistant-generated replies rather than on user text).

    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
    SG
    -0.07
    _sync
    -0.06
    THOOK
    -0.06
    ^^^^
    -0.06
    تين
    -0.06
    currentIndex
    -0.06
     Bukkit
    -0.06
     Пред
    -0.06
    -slot
    -0.06
    ska
    -0.06
    POSITIVE LOGITS
     Catalonia
    0.07
     salt
    0.06
    ーテ
    0.06
    (types
    0.06
    DataMember
    0.06
     lạnh
    0.06
    asad
    0.06
     discussions
    0.06
    áb
    0.06
    aday
    0.06
    Activations Density 0.015%

    No Known Activations