© 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. 90634
    Prev
    Next
    INDEX
    Explanations

    ively

    np_max-act · gemini-2.0-flash

    sentences discussing legal appeals and their outcomes.

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

    This neuron is keyed to adverbs ending in “-ly,” especially legal/technical adverbs (e.g., procedurally, substantively).

    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
     증가
    -0.06
     Alignment
    -0.06
    simulate
    -0.06
    iangle
    -0.06
     poc
    -0.06
     Hem
    -0.06
    ()<
    -0.06
     decreased
    -0.05
     electrode
    -0.05
    ZX
    -0.05
    POSITIVE LOGITS
    .getBoundingClientRect
    0.07
     standalone
    0.07
    undreds
    0.07
    aph
    0.07
    ionario
    0.06
     suyu
    0.06
     sheer
    0.06
    ск
    0.06
     наб
    0.06
    langle
    0.06
    Activations Density 0.001%

    No Known Activations