© 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. 126832
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens that occur in Markdown-style image/link URLs—specifically the journal‐citation strings (abbreviations like “med,” “phys,” “lond,” and numeric identifiers) inside the parentheses of `![](...)`.

    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
    international
    -0.07
    uxe
    -0.06
     ucwords
    -0.06
     aprox
    -0.06
    .radius
    -0.06
    Naz
    -0.06
     Classifier
    -0.06
     Hawth
    -0.06
     synchronize
    -0.06
     Europeans
    -0.06
    POSITIVE LOGITS
    #__
    0.06
    ска
    0.06
    amespace
    0.06
    unge
    0.06
     мали
    0.06
    AUSE
    0.06
    illum
    0.06
    ensible
    0.06
     incompet
    0.06
     fö
    0.06
    Activations Density 0.006%

    No Known Activations