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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 385
    Prev
    Next
    INDEX
    Explanations

    Text snippets

    np_max-act · gemini-2.0-flash

    The neuron is especially sensitive to narrative expressions of personal intent or feeling—words that convey someone’s thoughts, desires, or emotional state.

    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
    Quiet
    -0.07
    �
    -0.06
    .StObject
    -0.06
     روسیه
    -0.06
     thải
    -0.06
    ورش
    -0.06
     Johannesburg
    -0.06
    .dom
    -0.06
    structuring
    -0.06
    VISION
    -0.06
    POSITIVE LOGITS
     μ
    0.07
    πέ
    0.06
    .Enqueue
    0.06
     onCreateView
    0.06
    (private
    0.06
     теор
    0.06
    _Value
    0.06
    hosts
    0.06
     �
    0.06
    _specific
    0.06
    Activations Density 0.103%

    No Known Activations