© 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. 91563
    Prev
    Next
    INDEX
    Explanations

    grounds

    np_max-act · gemini-2.0-flash

    The neuron activates on instances of the legal term “grounds,” as in “on the grounds that,” marking where reasons or bases for motions or arguments are stated.

    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
     بیش
    -0.07
     Sek
    -0.07
     responseBody
    -0.07
    社会
    -0.06
    _ASS
    -0.06
     다시
    -0.06
    responses
    -0.06
     Clone
    -0.06
    _confirmation
    -0.06
     Veronica
    -0.06
    POSITIVE LOGITS
     grounds
    0.09
     GK
    0.07
    дина
    0.07
     bleed
    0.07
    	throw
    0.06
     alleges
    0.06
    μπ
    0.06
     Lakers
    0.06
    б
    0.06
     scoreboard
    0.06
    Activations Density 0.003%

    No Known Activations