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. 129045
    Prev
    Next
    INDEX
    Explanations

    injury

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of injuries (e.g., “injury,” body part + injury, injury‐related terms).

    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
    ovich
    -0.06
     market
    -0.06
     evaluations
    -0.06
    Lite
    -0.06
     FileOutputStream
    -0.06
     Market
    -0.06
    TIM
    -0.06
    slug
    -0.06
    �
    -0.06
    Falsy
    -0.06
    POSITIVE LOGITS
    strap
    0.06
     HttpContext
    0.06
    CRY
    0.06
    BarButton
    0.06
    rement
    0.06
    ภาค
    0.06
    _JS
    0.06
    četně
    0.06
     आस
    0.06
    енд
    0.06
    Activations Density 0.005%

    No Known Activations