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. 92825
    Prev
    Next
    INDEX
    Explanations

    positive scenic adjectives

    np_max-act · gemini-2.0-flash

    This neuron fires on positive, descriptive adjectives (like “beautiful,” “stunning,” “amazing”) and related words (“views,” “enjoy,” “offers”) used to praise scenic or impressive sights.

    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
    .Reflection
    -0.08
    _menus
    -0.07
    .exc
    -0.07
     heiß
    -0.06
     territorial
    -0.06
     misunderstood
    -0.06
    -ahead
    -0.06
    ीमत
    -0.06
     Bitcoins
    -0.06
    оч
    -0.06
    POSITIVE LOGITS
     ^↵
    0.06
    vail
    0.06
    printStats
    0.06
    key
    0.06
    setMessage
    0.06
    iteral
    0.06
    anggan
    0.06
    richText
    0.06
    :m
    0.06
    -entity
    0.06
    Activations Density 0.021%

    No Known Activations