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. 130350
    Prev
    Next
    INDEX
    Explanations

    /goto

    np_max-act · gemini-2.0-flash

    The neuron activates on the API‐documentation hyperlink URLs (e.g. the “http://docs.aws.amazon.com/goto/WebAPI/…” links) in code comments.

    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
    crime
    -0.07
    �
    -0.07
     перел
    -0.07
     pump
    -0.07
    組
    -0.07
    :UIControl
    -0.06
     Plates
    -0.06
    歩
    -0.06
    .ly
    -0.06
    	Il
    -0.06
    POSITIVE LOGITS
    /goto
    0.08
    �
    0.07
     olduğuna
    0.07
     SA
    0.07
    ayette
    0.06
     farklı
    0.06
    ernals
    0.06
     Vista
    0.06
     foundation
    0.06
     Nashville
    0.06
    Activations Density 0.000%

    No Known Activations