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. 1114
    Prev
    Next
    INDEX
    Explanations

    news articles

    np_max-act · gemini-2.0-flash

    This neuron activates on the initial words of news story leads or headlines, spotting tokens in the opening sentence of an article.

    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
    timestamp
    -0.07
    .proj
    -0.06
     WHETHER
    -0.06
    utory
    -0.06
    	items
    -0.06
    -input
    -0.06
     antlr
    -0.06
     styling
    -0.06
    backend
    -0.06
    盖
    -0.06
    POSITIVE LOGITS
    scientific
    0.07
     fname
    0.06
     confined
    0.06
    действ
    0.06
    Fight
    0.06
    drawing
    0.06
     <<-
    0.06
    ensch
    0.06
     forces
    0.06
    (statement
    0.06
    Activations Density 0.077%

    No Known Activations