© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 8759
    Prev
    Next
    INDEX
    Explanations

    attention

    np_max-act · gemini-2.0-flash

    The neuron activates on words and phrases that signal directing attention or focus (e.g. “focus,” “attention,” “directed,” “focused”).

    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
     denen
    -0.06
    ngx
    -0.06
    _BOOT
    -0.06
    normalized
    -0.06
     dominated
    -0.06
     conservative
    -0.06
     butterknife
    -0.06
     leftist
    -0.06
    AffineTransform
    -0.06
     surfing
    -0.06
    POSITIVE LOGITS
     Veranst
    0.07
     文
    0.07
    Destination
    0.06
    kt
    0.06
     діяльність
    0.06
    ```
    0.06
    лас
    0.06
    _measurement
    0.06
     Công
    0.06
     мяг
    0.06
    Activations Density 0.043%

    No Known Activations