© 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. 7810
    Prev
    Next
    INDEX
    Explanations

    Viewer discretion advised

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens from content‐warning advisory phrases (e.g. “discretion,” “advised”).

    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
    enda
    -0.08
    yaml
    -0.07
    ches
    -0.07
    (void
    -0.06
    .ax
    -0.06
     Sie
    -0.06
    igenous
    -0.06
    Fu
    -0.06
     Пра
    -0.06
     Hind
    -0.06
    POSITIVE LOGITS
     disclosures
    0.07
     вв
    0.07
    ้ง
    0.06
     vriend
    0.06
    .getSeconds
    0.06
    孩
    0.06
     сим
    0.06
    血
    0.06
     observation
    0.06
    ITION
    0.06
    Activations Density 0.002%

    No Known Activations