© 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. 92613
    Prev
    Next
    INDEX
    Explanations

    gay

    np_max-act · gemini-2.0-flash

    The neuron activates on explicit sexual‐orientation and gender‐identity labels (e.g. LGBT, LGBTQ, lesbian, gay, bisexual, transgender, queer).

    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
     Burlington
    -0.07
    研
    -0.07
     】
    -0.07
    ña
    -0.06
    west
    -0.06
    -------------</
    -0.06
     giáo
    -0.06
    Rock
    -0.06
    Te
    -0.06
    (api
    -0.06
    POSITIVE LOGITS
    .raises
    0.08
    mentioned
    0.07
     destination
    0.07
     جر
    0.07
     PTS
    0.06
     tamp
    0.06
    -turned
    0.06
     instantiate
    0.06
     expansive
    0.06
    indrical
    0.06
    Activations Density 0.006%

    No Known Activations