© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 91934
    Prev
    Next
    INDEX
    Explanations

    kissing

    np_max-act · gemini-2.0-flash

    The neuron responds to the special conversation‐control markers and metadata tokens (e.g. “<|start_header_id|>”, “<|end_header_id|>”, user/assistant labels, etc.).

    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
     AFL
    -0.07
    оре
    -0.07
    ştir
    -0.06
     love
    -0.06
    lore
    -0.06
    KHR
    -0.06
    ificantly
    -0.06
    ALLOC
    -0.06
    \Event
    -0.06
    Fil
    -0.06
    POSITIVE LOGITS
    (filename
    0.07
     Αρχ
    0.07
     제출
    0.06
    	f
    0.06
     usual
    0.06
    ToFile
    0.06
     unk
    0.06
     sơ
    0.06
    Clients
    0.06
     veriler
    0.06
    Activations Density 0.019%

    No Known Activations