© 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. 90885
    Prev
    Next
    INDEX
    Explanations

    Single letters

    np_max-act · gemini-2.0-flash

    The neuron fires on parts of hyphenated acronyms or proper names (e.g. model or product codes) that include a capital‐letter sequence joined by a dash.

    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
     стара
    -0.08
    ец
    -0.07
     Demp
    -0.07
    енка
    -0.06
    陰
    -0.06
     přeb
    -0.06
    ولا
    -0.06
    xAA
    -0.06
     akci
    -0.06
    ubb
    -0.06
    POSITIVE LOGITS
    -t
    0.08
    -s
    0.08
    -S
    0.08
    -N
    0.07
    -d
    0.07
    -l
    0.07
    -w
    0.07
    -R
    0.07
    -g
    0.07
    "S
    0.07
    Activations Density 0.063%

    No Known Activations