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. 1533
    Prev
    Next
    INDEX
    Explanations

    beaches

    np_max-act · gemini-2.0-flash

    The neuron responds to French-language tokens, effectively identifying when the text is in French.

    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
    iddles
    -0.07
    incer
    -0.07
    	search
    -0.07
    .keys
    -0.07
                      
    -0.07
    Hunter
    -0.07
    mother
    -0.07
          
    -0.07
     Лени
    -0.06
    cura
    -0.06
    POSITIVE LOGITS
     있을
    0.06
     Burg
    0.06
     bounty
    0.06
     estoy
    0.06
    insert
    0.06
    _Insert
    0.06
     ill
    0.06
     ві
    0.06
     прин
    0.06
     kaç
    0.06
    Activations Density 0.035%

    No Known Activations