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. 86213
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    The neuron fires on the common email-opening phrase “I am writing to …,” flagging those tokens in the message’s introduction.

    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
    eous
    -0.06
    -handle
    -0.06
     (/
    -0.06
     }}}
    -0.06
    )new
    -0.06
    (core
    -0.06
     ici
    -0.06
     ngOnInit
    -0.06
    妻
    -0.06
     variations
    -0.06
    POSITIVE LOGITS
    /go
    0.06
    esses
    0.06
    izes
    0.06
     Merchant
    0.06
    urvey
    0.06
     Instit
    0.06
     distr
    0.06
     โดย
    0.06
    gas
    0.06
    Fitness
    0.06
    Activations Density 0.013%

    No Known Activations