© 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. 68948
    Prev
    Next
    INDEX
    Explanations

    Sports field location

    np_max-act · gemini-2.0-flash

    This neuron responds to words that describe spatial distance in sports reporting—especially terms like “outside,” “close,” or “range” that specify where a shot or play occurred relative to the goal.

    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
     running
    -0.06
     allowance
    -0.06
    ━━━━
    -0.06
    壊
    -0.06
     Hate
    -0.06
     Vk
    -0.05
    Speaker
    -0.05
     extradition
    -0.05
     능
    -0.05
    component
    -0.05
    POSITIVE LOGITS
    .Top
    0.07
    happy
    0.07
    _frm
    0.07
     fos
    0.07
    Zend
    0.07
    YEAR
    0.07
    ناد
    0.07
    /general
    0.06
     Свят
    0.06
     öden
    0.06
    Activations Density 0.004%

    No Known Activations