© 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. 73545
    Prev
    Next
    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    This neuron detects type names and punctuation within angle-bracketed generic type parameter lists (e.g., the contents of “<…>” in Java generics).

    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
     />,
    -0.07
     dataSet
    -0.07
     sebep
    -0.06
    えば
    -0.06
    りに
    -0.06
    ];//
    -0.06
    serialize
    -0.06
    　　 　
    -0.06
    金属
    -0.06
     sympt
    -0.06
    POSITIVE LOGITS
     Aug
    0.07
     Boat
    0.07
     Paid
    0.06
     eag
    0.06
     paid
    0.06
     Pg
    0.06
    _ck
    0.06
    (avg
    0.06
    -clean
    0.06
     ổ
    0.06
    Activations Density 0.020%

    No Known Activations