© 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. 9025
    Prev
    Next
    INDEX
    Explanations

    online production

    np_max-act · gemini-2.0-flash

    This neuron detects floating-point numbers (like “5.70703125”, “0.822265625”, etc.) that appear in the document metadata or formatting headers/footers.

    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
     ogr
    -0.07
    ël
    -0.07
     Jung
    -0.07
    enn
    -0.07
     relaxing
    -0.06
     "#"
    -0.06
    etim
    -0.06
     Nim
    -0.06
    _COLUMNS
    -0.06
     Pickup
    -0.06
    POSITIVE LOGITS
    ↵↵↵↵
    0.08
    ernet
    0.07
    ↵↵↵↵↵↵↵↵↵↵↵↵
    0.07
     конс
    0.07
     checker
    0.06
    		
↵
↵
    0.06
    ования
    0.06
    [sub
    0.06
    ↵↵↵
    0.06
    eee
    0.06
    Activations Density 0.004%

    No Known Activations