© 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. 110098
    Prev
    Next
    INDEX
    Explanations

    RX serialization

    np_max-act · gemini-2.0-flash

    technical discussions about JSON serialization and deserialization configurations.

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

    The neuron lights up on tokens referring to data “reading” or conversion operations—e.g. reads/reading entities, DeserializationFeature, mapToObject, etc.—i.e. mentions of data‐retrieval and object‐mapping/deserialization steps.

    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
    PLICATION
    -0.07
    ีซ
    -0.07
    pr
    -0.07
    ipp
    -0.07
    -network
    -0.06
    paste
    -0.06
    _end
    -0.06
     ود
    -0.06
    {s
    -0.06
     HT
    -0.06
    POSITIVE LOGITS
     إليه
    0.07
     požadav
    0.07
     etmiş
    0.07
     damer
    0.07
     образом
    0.06
    testimonial
    0.06
    แทน
    0.06
     حی
    0.06
    InstantiationException
    0.06
    portrait
    0.06
    Activations Density 0.026%

    No Known Activations