© 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. 110311
    Prev
    Next
    INDEX
    Explanations

    (

    np_max-act · gemini-2.0-flash

    This neuron fires on occurrences of technical term expansions immediately followed by their abbreviations in parentheses.

    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
    іг
    -0.07
     cure
    -0.07
    -olds
    -0.06
    rzy
    -0.06
     getUsers
    -0.06
    щество
    -0.06
    BALL
    -0.06
     вариант
    -0.06
    quiz
    -0.06
    atoire
    -0.06
    POSITIVE LOGITS
     clearfix
    0.08
    既
    0.07
    (handles
    0.07
     ByteArrayInputStream
    0.07
     credible
    0.06
     Gathering
    0.06
    .dex
    0.06
     Emails
    0.06
     tedbir
    0.06
     Batman
    0.06
    Activations Density 0.575%

    No Known Activations