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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 90116
    Prev
    Next
    INDEX
    Explanations

    //

    np_max-act · gemini-2.0-flash

    phrases related to software licensing and legal documentation.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron strongly fires on ASCII-art code comment separators (long runs of “//===…===” or “//------…” lines that act as section dividers).

    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
     dile
    -0.07
     office
    -0.07
     Cleaning
    -0.07
    titles
    -0.06
     cleaning
    -0.06
    _repo
    -0.06
    -La
    -0.06
     either
    -0.06
     στις
    -0.06
     недели
    -0.06
    POSITIVE LOGITS
    Lastly
    0.07
    (nc
    0.07
     responsable
    0.06
     Fourth
    0.06
    ्षमत
    0.06
     Nonetheless
    0.06
     `↵
    0.06
     concentrating
    0.06
     رئیس
    0.06
     مفهوم
    0.06
    Activations Density 0.004%

    No Known Activations