© 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. 25361
    Prev
    Next
    INDEX
    Explanations

    chloride

    np_max-act · gemini-2.0-flash

    The neuron consistently activates on instances of “chloride” (especially in catalyst names like aluminum chloride or ferric chloride), indicating it’s detecting that word.

    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
     όπως
    -0.07
     scrollTo
    -0.06
    .GONE
    -0.06
    CppMethodPointer
    -0.06
    _OBJ
    -0.06
     scrollTop
    -0.06
    阳
    -0.06
     hatch
    -0.06
    Atl
    -0.06
    879
    -0.06
    POSITIVE LOGITS
     nationalists
    0.07
     retrieving
    0.06
    غال
    0.06
    ़ों
    0.06
    _Core
    0.06
     Morav
    0.06
    Courier
    0.06
    υμ
    0.06
    {_
    0.06
    chied
    0.06
    Activations Density 0.003%

    No Known Activations