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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 128925
    Prev
    Next
    INDEX
    Explanations

    numerical values

    np_max-act · gemini-2.0-flash

    The neuron selectively responds to numeric tokens (quantities, percentages, dates, and other figures) in the text.

    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
    _attention
    -0.07
    -first
    -0.06
     advisory
    -0.06
    ivan
    -0.06
    ียง
    -0.06
    pecial
    -0.06
     ringing
    -0.06
    quared
    -0.06
    _disable
    -0.06
    Bid
    -0.06
    POSITIVE LOGITS
     повинні
    0.07
    	↵	↵	↵	↵
    0.07
    يط
    0.07
     würde
    0.06
    	DBG
    0.06
     ขาย
    0.06
     diye
    0.06
    	var
    0.06
    ATIVE
    0.06
    uyordu
    0.06
    Activations Density 0.047%

    No Known Activations