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. 81285
    Prev
    Next
    INDEX
    Explanations

    information

    np_max-act · gemini-2.0-flash

    phrases related to technical instructions, particularly where settings or configurations are explained.

    oai_token-act-pair · gpt-4oTriggered by @tcai

    The neuron detects specific configuration or UI label names (e.g. menu options and setting names) in technical instruction text.

    oai_token-act-pair · o4-miniTriggered by @vishnuveerathu

    or

    np_max-act · o4-miniTriggered by @vishnuveerathu

    or

    np_max-act-logits · o4-miniTriggered by @vishnuveerathu

    technical instructions related to computer and printer management.

    oai_token-act-pair · gpt-4o-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
    ตลอด
    -0.06
    ]==
    -0.06
     Aleppo
    -0.06
    -addon
    -0.06
    农
    -0.06
    unix
    -0.06
     rowCount
    -0.06
     quickest
    -0.06
     Marian
    -0.06
    ковые
    -0.06
    POSITIVE LOGITS
     RET
    0.07
     Methodist
    0.07
    、「
    0.07
     توسعه
    0.06
    	logging
    0.06
    .Child
    0.06
     Come
    0.06
     Secretary
    0.06
     DISTINCT
    0.06
     troubleshooting
    0.06
    Activations Density 0.008%

    No Known Activations