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. 104383
    Prev
    Next
    INDEX
    Explanations

    what

    np_max-act · gemini-2.0-flash

    conversational phrases and interactions related to customer service scenarios.

    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
    _SPE
    -0.07
    Away
    -0.06
    Pretty
    -0.06
    _BREAK
    -0.06
    -key
    -0.06
    -scalable
    -0.06
                                                 
    -0.06
    альную
    -0.06
     двух
    -0.06
    やす
    -0.06
    POSITIVE LOGITS
     Medicaid
    0.07
     συμπ
    0.07
     disagreements
    0.07
    ंडल
    0.06
    ment
    0.06
     revival
    0.06
    _property
    0.06
     신
    0.06
    .addField
    0.06
     <
    0.06
    Activations Density 0.028%

    No Known Activations