© 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
    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. 92617
    Prev
    Next
    INDEX
    Explanations

    contrarian; doing the opposite

    np_max-act · gemini-2.0-flash

    The neuron strongly responds to mentions of defying or going against OpenAI’s policies and guidelines.

    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
     monk
    -0.07
    _a
    -0.07
     срав
    -0.06
     tote
    -0.06
    .__
    -0.06
     testcase
    -0.06
     Editors
    -0.06
    .constraint
    -0.06
    	w
    -0.06
    .HorizontalAlignment
    -0.06
    POSITIVE LOGITS
     Coronavirus
    0.07
    134
    0.07
     Spotify
    0.07
     rising
    0.07
     Ghost
    0.06
     ali
    0.06
     chol
    0.06
     Carb
    0.06
     určitě
    0.06
     marked
    0.06
    Activations Density 0.006%

    No Known Activations