© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 27608
    Prev
    Next
    INDEX
    Explanations

    he

    np_max-act · gemini-2.0-flash

    main narrative/action verbs (tokens that mark events or actions in a story, especially past-tense auxiliaries and verbs).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    повід
    -0.07
    Statement
    -0.07
     unspecified
    -0.06
    	param
    -0.06
    ازی
    -0.06
    _usuario
    -0.06
    notations
    -0.06
     YT
    -0.06
    Generating
    -0.06
     Dah
    -0.06
    POSITIVE LOGITS
     سي
    0.07
    ูล
    0.06
     enim
    0.06
    竞
    0.06
    هم
    0.05
     korun
    0.05
     HAPP
    0.05
     DID
    0.05
    حه
    0.05
    /photos
    0.05
    Activations Density 0.119%

    No Known Activations