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. 84795
    Prev
    Next
    INDEX
    Explanations

    conversational writing

    np_max-act · gemini-2.0-flash

    mentions of race in discussions about social behavior or attitudes.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron detects the comma immediately following a conditional demographic clause (e.g. the “If you’re a black person,” comma).

    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
    �
    -0.06
     hypoc
    -0.06
     Covenant
    -0.06
     conce
    -0.06
    	ap
    -0.06
    анти
    -0.06
     workspace
    -0.06
     대부분
    -0.06
    ndef
    -0.06
    ap
    -0.06
    POSITIVE LOGITS
    .music
    0.07
    shared
    0.07
    _BIT
    0.07
     hotline
    0.07
     ster
    0.07
    (red
    0.06
     derby
    0.06
     Ül
    0.06
    :::::::::::::
    0.06
     Micha
    0.06
    Activations Density 0.003%

    No Known Activations