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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 85497
    Prev
    Next
    INDEX
    Explanations

    data pattern analysis

    np_max-act · gemini-2.0-flash

    technical terms and concepts related to AI and its applications.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron detects technical data‐analysis and processing terminology, activating on words like “data,” “analysis,” “identify,” “patterns,” and “trends.”

    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
     wa
    -0.07
    516
    -0.07
    ालय
    -0.06
    ikh
    -0.06
     Numbers
    -0.06
    tdown
    -0.06
     Duke
    -0.06
     McDon
    -0.06
    وک
    -0.06
     PKK
    -0.06
    POSITIVE LOGITS
    екти
    0.07
    -unstyled
    0.07
    เ�
    0.06
    							  
    0.06
     çalışma
    0.06
     alanı
    0.06
     �
    0.06
     اعتر
    0.06
    .flash
    0.06
    ály
    0.06
    Activations Density 0.031%

    No Known Activations