© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 14026
    Prev
    Next
    INDEX
    Explanations

    directions

    np_max-act · gemini-2.0-flash

    The neuron activates on compass‐direction or orientation phrases (e.g. “to the west,” “northeastern,” “on the east flank,” etc.), i.e. tokens marking relative geographic direction/location.

    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
    ovní
    -0.07
    bru
    -0.07
    (order
    -0.07
     stopwords
    -0.06
    iks
    -0.06
     Vision
    -0.06
     wäre
    -0.06
     detain
    -0.06
    .EOF
    -0.06
     hoops
    -0.06
    POSITIVE LOGITS
     Получ
    0.06
    ()['
    0.06
    RequestBody
    0.06
    	token
    0.06
     impress
    0.06
    _('
    0.06
    DataTask
    0.06
    Spe
    0.06
    .'
    0.06
    โ
    0.06
    Activations Density 0.014%

    No Known Activations