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. 329
    Prev
    Next
    INDEX
    Explanations

    diverse text/conversations

    np_max-act · gemini-2.0-flash

    elements related to planning and organizing course content.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on self-descriptions of one’s professional role or background (e.g. “I’m a historian by training,” “I’m a nutrition therapist”).

    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
     změ
    -0.07
    няя
    -0.06
     Buffy
    -0.06
     Національ
    -0.06
     Cly
    -0.06
    .cgColor
    -0.06
    比較
    -0.06
    (test
    -0.06
     أنا
    -0.06
    .sb
    -0.05
    POSITIVE LOGITS
     ‐
    0.07
    ).</
    0.07
    インタ
    0.07
    _required
    0.07
    shortcut
    0.07
    _DEPEND
    0.07
    ?>><?
    0.06
    Insurance
    0.06
    Pull
    0.06
    rotein
    0.06
    Activations Density 0.030%

    No Known Activations