© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 42701
    Prev
    Next
    INDEX
    Explanations

    Wanting/Planning

    np_max-act · gemini-2.0-flash

    specific legal or medical claims related to negligence and misdiagnosis.

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

    The neuron activates on words that express someone’s intention or desire (e.g. “intention,” “desire,” “intent,” “expressed intent”).

    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
    acc
    -0.07
    public
    -0.06
    genes
    -0.06
    019
    -0.06
    ’h
    -0.06
    formation
    -0.06
    select
    -0.06
     interrog
    -0.06
     بدان
    -0.06
    orce
    -0.06
    POSITIVE LOGITS
     unab
    0.07
     :-↵
    0.06
     которым
    0.06
     зай
    0.06
     управ
    0.06
     egy
    0.06
     odst
    0.06
     API
    0.06
     verm
    0.06
    ，被
    0.06
    Activations Density 0.029%

    No Known Activations