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. 118210
    Prev
    Next
    INDEX
    Explanations

    do

    np_max-act · gemini-2.0-flash

    The main thing this neuron does is pick out instances of the verb “do.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    code snippets related to method definitions and object-oriented programming concepts.

    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
     welt
    -0.08
    arius
    -0.07
     Fransız
    -0.07
    ]]></
    -0.06
    راف
    -0.06
    Salt
    -0.06
    exter
    -0.06
     Xavier
    -0.06
     fears
    -0.06
     cavern
    -0.06
    POSITIVE LOGITS
     do
    0.16
     DO
    0.14
    do
    0.12
     doing
    0.12
     Do
    0.12
    DO
    0.11
    Do
    0.11
    	do
    0.11
    -do
    0.11
     Doing
    0.11
    Activations Density 0.189%

    No Known Activations