© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 91498
    Prev
    Next
    INDEX
    Explanations

    scientific/technical texts

    np_max-act · gemini-2.0-flash

    This neuron responds to scientific‐jargon terms carrying the “-ine” or “-ria” suffix—e.g. chemical names like melamine and organelle names like mitochondria.

    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
     malt
    -0.07
    .Main
    -0.06
    _tri
    -0.06
     faded
    -0.06
     hvor
    -0.06
     acoustic
    -0.06
     dining
    -0.06
     достаточно
    -0.06
    setAttribute
    -0.06
     всё
    -0.06
    POSITIVE LOGITS
     Enums
    0.07
    彦
    0.07
    createTime
    0.06
    /package
    0.06
     کسی
    0.06
    @class
    0.06
     알아
    0.06
     کننده
    0.06
    AVOR
    0.06
     gcc
    0.06
    Activations Density 0.202%

    No Known Activations