© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 12290
    Prev
    Next
    INDEX
    Explanations

    articles

    np_max-act · gemini-2.0-flash

    This neuron detects general academic “background” or “prior work” phrases that introduce established facts or summarize earlier research (e.g. It is known that…, Since…, Much effort has…).

    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
     Evan
    -0.06
     Coinbase
    -0.06
    alis
    -0.06
     mActivity
    -0.06
     frail
    -0.06
     containers
    -0.06
    Unhandled
    -0.06
     coolant
    -0.06
    mod
    -0.06
     Clients
    -0.06
    POSITIVE LOGITS
    	mutex
    0.08
    ستانی
    0.07
     padded
    0.07
     continual
    0.07
    _UNIQUE
    0.07
     olun
    0.07
     ξ
    0.06
    _pic
    0.06
    (rng
    0.06
    许多
    0.06
    Activations Density 0.024%

    No Known Activations