© 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. 125070
    Prev
    Next
    INDEX
    Explanations

    rebuild

    np_max-act · gemini-2.0-flash

    The neuron activates on terms referring to rebuilding or recovery efforts (e.g., “rebuilding,” “recovery,” “rebuild,” “rebuilding”).

    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
     dogs
    -0.06
    χα
    -0.06
     hardware
    -0.06
    creating
    -0.06
    Flash
    -0.06
     ioctl
    -0.06
     Bent
    -0.06
     Creed
    -0.06
    โร
    -0.06
    iples
    -0.06
    POSITIVE LOGITS
    Carrier
    0.07
     Carrier
    0.06
     sexes
    0.06
     held
    0.06
    ))
↵
    0.06
    mg
    0.06
    illet
    0.06
    vertiser
    0.06
     forehead
    0.06
     ifstream
    0.06
    Activations Density 0.019%

    No Known Activations