© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 41428
    Prev
    Next
    INDEX
    Explanations

    coconut

    np_max-act · gemini-2.0-flash

    The neuron peaks on mentions of specific technologies, frameworks, or product names (e.g. Triton, hydroSMART, Python, Svelte, Chrome). It signals when spotting these proper‐noun tech/tool identifiers.

    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
     Shel
    -0.06
    Official
    -0.06
     بسیاری
    -0.06
    goods
    -0.06
    (Field
    -0.06
    -edge
    -0.06
    IVA
    -0.06
    istros
    -0.06
     ammunition
    -0.06
    .encrypt
    -0.06
    POSITIVE LOGITS
    							↵
    0.08
     良
    0.07
     dataGridView
    0.07
     rot
    0.07
    .bottomAnchor
    0.07
    fontWeight
    0.07
     Csv
    0.07
     banners
    0.07
     Rot
    0.06
    NES
    0.06
    Activations Density 0.095%

    No Known Activations