© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 73747
    Prev
    Next
    INDEX
    Explanations

    Hexadecimal code

    np_max-act · gemini-2.0-flash

    lines of code or comments in programming syntax, particularly pertaining to character encoding or mapping.

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

    The neuron is tuned to pick up on C‐style comment delimiters and license header blocks (the “/* … */” sections at the top of source files).

    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
    Associated
    -0.07
     aggressively
    -0.06
    uelve
    -0.06
     Supplements
    -0.06
     remin
    -0.06
     setups
    -0.06
    ẽ
    -0.06
     datetime
    -0.06
     Rep
    -0.06
     dtype
    -0.06
    POSITIVE LOGITS
     boredom
    0.06
    cro
    0.06
    eks
    0.06
    	parse
    0.06
     filtered
    0.06
     دریافت
    0.06
    ISTA
    0.06
     compliant
    0.05
    =_('
    0.05
     WARRANTIES
    0.05
    Activations Density 0.001%

    No Known Activations