© 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. 92193
    Prev
    Next
    INDEX
    Explanations

    Forum/online discussions

    np_max-act · gemini-2.0-flash

    This neuron activates on forum user handles (the author names in “Originally Posted by …” headers). It’s detecting parts of usernames.

    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
     "".
    -0.07
    iel
    -0.07
    LIK
    -0.06
     поба
    -0.06
     hamstring
    -0.06
    .picture
    -0.06
     SDLK
    -0.06
     sample
    -0.06
    ORN
    -0.06
    _clicked
    -0.06
    POSITIVE LOGITS
    vlan
    0.06
     solar
    0.06
     innovative
    0.06
    ={()=>
    0.06
    ůvodu
    0.06
    Solar
    0.06
    抵
    0.05
    continent
    0.05
    昨
    0.05
    宝
    0.05
    Activations Density 0.014%

    No Known Activations