© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 52714
    Prev
    Next
    INDEX
    Explanations

    sentence continuations

    np_max-act · gemini-2.0-flash

    This neuron responds to discourse-structuring cues—transitional and signposting elements that authors use to move between points, give examples, or frame arguments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    critical commentary on government actions and decisions.

    oai_token-act-pair · gpt-4o-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
    او
    -0.07
    hart
    -0.07
    ´t
    -0.07
     elaborate
    -0.06
     student
    -0.06
     cart
    -0.06
     department
    -0.06
    (Double
    -0.06
    verted
    -0.06
     bufio
    -0.06
    POSITIVE LOGITS
    ån
    0.07
    asının
    0.07
    llu
    0.06
    .Ε
    0.06
    elerle
    0.06
     Indicator
    0.06
     dışında
    0.06
    ültür
    0.06
     zz
    0.06
    Tên
    0.06
    Activations Density 0.286%

    No Known Activations