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. 86247
    Prev
    Next
    INDEX
    Explanations

    Summations and math variables

    np_max-act · gemini-2.0-flash

    This neuron primarily detects C/C++ preprocessor “#include” directives.

    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
    -b
    -0.07
    OP
    -0.07
    界
    -0.07
    控
    -0.06
     dür
    -0.06
    شور
    -0.06
    換
    -0.06
     stakeholders
    -0.06
     lateinit
    -0.06
    etim
    -0.06
    POSITIVE LOGITS
     [<
    0.08
     specifies
    0.07
    とな
    0.06
     brings
    0.06
    (problem
    0.06
    そして
    0.06
     Làm
    0.06
    UCCEEDED
    0.06
    _caption
    0.06
     Carb
    0.06
    Activations Density 0.018%

    No Known Activations