© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 13598
    Prev
    Next
    INDEX
    Explanations

    brackets

    np_max-act · gemini-2.0-flash

    The neuron detects the literal “[" bracket token (used to mark metadata or section-header annotations, e.g. “[Instruction]” or “[rules]”).

    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
    worked
    -0.06
    تون
    -0.06
    �
    -0.06
    Phones
    -0.06
     '':↵
    -0.06
    DN
    -0.06
    iform
    -0.06
    configured
    -0.06
    }]
    -0.06
     iface
    -0.06
    POSITIVE LOGITS
    (parseInt
    0.08
     FOOT
    0.07
     krij
    0.07
     fitness
    0.07
    (prom
    0.07
     UIAlertController
    0.07
    SOFTWARE
    0.07
    潜
    0.07
    ُوا
    0.06
     uncont
    0.06
    Activations Density 0.020%

    No Known Activations