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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 85632
    Prev
    Next
    INDEX
    Explanations

    Targeting "you"

    np_max-act · gemini-2.0-flash

    The neuron detects marketing call-to-action phrases aimed at the reader (e.g. “this is for you,” “then this course is for you,” etc.).

    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
     pissed
    -0.07
        	 
    -0.07
    	dist
    -0.07
    ош
    -0.06
     temporary
    -0.06
     assistir
    -0.06
     Levin
    -0.06
    partial
    -0.06
    串
    -0.06
     lifting
    -0.06
    POSITIVE LOGITS
    літ
    0.07
    emarks
    0.07
     Sony
    0.06
    >Edit
    0.06
     결혼
    0.06
     financially
    0.06
     Allows
    0.06
    ΗΜ
    0.06
     Minutes
    0.06
     Panasonic
    0.06
    Activations Density 0.059%

    No Known Activations