© 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. 13032
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron detects special structural or metadata tokens (e.g. “<|begin_of_text|>”, “<|eot_id|>”, header markers) used for formatting the conversation.

    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
     تهران
    -0.06
    ilerek
    -0.06
    (PORT
    -0.06
    ().↵
    -0.06
     aluno
    -0.06
     playerName
    -0.06
    :test
    -0.06
    عن
    -0.06
    造成
    -0.06
    曾
    -0.06
    POSITIVE LOGITS
    _LD
    0.08
     mathematics
    0.07
     puzzles
    0.07
    .AutoScaleDimensions
    0.06
     Scientists
    0.06
     qual
    0.06
    _med
    0.06
     ход
    0.06
     confronted
    0.06
     Marg
    0.06
    Activations Density 0.007%

    No Known Activations