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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 86053
    Prev
    Next
    INDEX
    Explanations

    geographic regions

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of geographical regions or continent names in the text.

    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
    enze
    -0.08
    Tech
    -0.07
    isk
    -0.07
     ceasefire
    -0.07
    SURE
    -0.06
    Reverse
    -0.06
     Tech
    -0.06
    artz
    -0.06
    odzi
    -0.06
    IP
    -0.06
    POSITIVE LOGITS
    bt
    0.07
     gracefully
    0.07
     endregion
    0.06
    =findViewById
    0.06
    neg
    0.06
     dağ
    0.06
    abetes
    0.06
    ³
    0.06
    ฟร
    0.06
     gerekiyor
    0.06
    Activations Density 0.033%

    No Known Activations