© 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. 872
    Prev
    Next
    INDEX
    Explanations

    -b

    np_max-act · gemini-2.0-flash

    This neuron detects IUPAC locant notation of the form “[numbers-b]” (e.g., the “-b” in bracketed positions like “[2,3-b]”) in chemical names.

    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
    具
    -0.06
    жив
    -0.06
    acid
    -0.06
    /comment
    -0.06
    its
    -0.06
    .leave
    -0.06
    .next
    -0.06
    'є
    -0.06
    	unit
    -0.06
    embers
    -0.06
    POSITIVE LOGITS
    (mx
    0.07
    CppClass
    0.06
     throwable
    0.06
     â
    0.06
     Lucia
    0.06
    Shar
    0.06
    _increase
    0.06
    ENSITY
    0.06
     очень
    0.06
     carnival
    0.06
    Activations Density 0.005%

    No Known Activations