© 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. 109436
    Prev
    Next
    INDEX
    Explanations

    Image captions

    np_max-act · gemini-2.0-flash

    This neuron activates on image‐caption metadata (words like “Photo:”, “Image” and the accompanying numbering/credit tokens).

    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
    芬
    -0.07
     jist
    -0.07
     yüzde
    -0.06
     histograms
    -0.06
     대부분
    -0.06
     Honduras
    -0.06
     confirmPassword
    -0.06
     الأك
    -0.06
     ducks
    -0.06
    格式
    -0.06
    POSITIVE LOGITS
     Willie
    0.06
    лі
    0.06
     ((
    0.06
    óln
    0.06
    vision
    0.06
     GAR
    0.06
    _clr
    0.06
    скому
    0.06
    ofday
    0.06
    leen
    0.06
    Activations Density 0.008%

    No Known Activations