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. 130767
    Prev
    Next
    INDEX
    Explanations

    Earning money

    np_max-act · gemini-2.0-flash

    The neuron detects imperative action verbs (commands) used to introduce suggestion list items.

    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
     detector
    -0.07
     nó
    -0.07
     mosque
    -0.07
    -number
    -0.06
     Ford
    -0.06
     tents
    -0.06
    Quotes
    -0.06
    ्त
    -0.06
    -goal
    -0.06
     detection
    -0.06
    POSITIVE LOGITS
     нес
    0.06
    andExpect
    0.06
    ảo
    0.06
    Fully
    0.06
     convertView
    0.06
     Glover
    0.06
    createClass
    0.06
     TNT
    0.06
    еди
    0.06
    =headers
    0.06
    Activations Density 0.030%

    No Known Activations