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. 86224
    Prev
    Next
    INDEX
    Explanations

    revenue

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on mentions of “revenue” (i.e., the word “revenue” and associated revenue figures).

    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
    同学
    -0.07
     použití
    -0.06
     Placeholder
    -0.06
    .gson
    -0.06
    áč
    -0.06
    ้าช
    -0.06
    禮
    -0.06
    .isnan
    -0.06
    .forRoot
    -0.06
     '_
    -0.06
    POSITIVE LOGITS
     revenue
    0.14
     Revenue
    0.11
     revenues
    0.10
    Revenue
    0.09
    venue
    0.08
    V
    0.08
    We
    0.08
     velocity
    0.08
     produkt
    0.08
    Nova
    0.08
    Activations Density 0.005%

    No Known Activations