© 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. 89851
    Prev
    Next
    INDEX
    Explanations

    image

    np_max-act · gemini-2.0-flash

    This neuron activates on occurrences of the word “image” (and its plural “images”).

    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
     slow
    -0.08
     pay
    -0.07
     nrows
    -0.07
     complicated
    -0.07
     Bacon
    -0.07
    bone
    -0.06
     Plum
    -0.06
    rong
    -0.06
    EXIST
    -0.06
     Psalm
    -0.06
    POSITIVE LOGITS
     images
    0.09
     image
    0.09
    名
    0.08
     likeness
    0.07
     Images
    0.07
     imagery
    0.07
     imagen
    0.07
    ตน
    0.07
    ----------------
    0.07
    Images
    0.07
    Activations Density 0.018%

    No Known Activations