© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 55456
    Prev
    Next
    INDEX
    Explanations

    programming code

    np_max-act · gemini-2.0-flash

    code snippets related to event handling in text input fields.

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

    The neuron fires on programming‐style syntax—i.e. it detects when the text is showing code (method calls, regex patterns, attributes, etc.).

    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
    geries
    -0.07
     RUS
    -0.07
    istogram
    -0.07
     aç
    -0.06
    fitness
    -0.06
     clases
    -0.06
    volume
    -0.06
    Restore
    -0.06
    iles
    -0.06
    алю
    -0.06
    POSITIVE LOGITS
    collector
    0.07
     campaigned
    0.07
     Collaboration
    0.06
     advertised
    0.06
    ίζ
    0.06
     Toronto
    0.06
    _comment
    0.06
     srov
    0.06
     pleading
    0.06
    �
    0.06
    Activations Density 0.026%

    No Known Activations