Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 84423
    Prev
    Next
    INDEX
    Explanations

    amusement

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of amusement-park attractions (e.g. parks, rides, roller coasters, water parks).

    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
    ars
    -0.07
     должны
    -0.06
     book
    -0.06
    field
    -0.06
    ijk
    -0.06
     structs
    -0.06
     shields
    -0.06
     tp
    -0.06
     static
    -0.06
    	Return
    -0.06
    POSITIVE LOGITS
     �
    0.07
     amusement
    0.07
    .toLowerCase
    0.06
     Georges
    0.06
     sweating
    0.06
    ічний
    0.06
    �습니다
    0.06
    .sources
    0.06
     Institution
    0.06
     همه
    0.06
    Activations Density 0.019%

    No Known Activations