© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 52216
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    phrases related to the performance and functionality of neural networks, particularly in the context of link construction and filtering strategies.

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

    The main thing this neuron does is detect sentence-initial discourse markers and transitional words (e.g. “However,” “In addition,” “Currently,” “Therefore,” 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
    late
    -0.07
     وج
    -0.07
    pane
    -0.07
    	pm
    -0.06
    Phys
    -0.06
    貨
    -0.06
    Pal
    -0.06
     sandy
    -0.06
    ош
    -0.06
    $max
    -0.06
    POSITIVE LOGITS
     Mayor
    0.07
    ('//*[@
    0.07
    idend
    0.06
    (平成
    0.06
    .SetToolTip
    0.06
    Caught
    0.06
     листопада
    0.06
    WindowTitle
    0.06
    osti
    0.06
    .ibm
    0.06
    Activations Density 0.035%

    No Known Activations