© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 92688
    Prev
    Next
    INDEX
    Explanations

    Chat/Conversations

    np_max-act · gemini-2.0-flash

    This neuron detects conversational greetings and salutations (tokens used to say “hello” or open a conversation).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     futuristic
    -0.08
    めた
    -0.07
     wurde
    -0.07
    iry
    -0.07
     McA
    -0.06
    itos
    -0.06
     strips
    -0.06
     trembling
    -0.06
    recursive
    -0.06
    .total
    -0.06
    POSITIVE LOGITS
     lodash
    0.07
     presets
    0.07
    _PANEL
    0.06
    ディース
    0.06
     นาง
    0.06
    Якщо
    0.06
    违
    0.06
    尿
    0.06
    علام
    0.06
    車
    0.06
    Activations Density 0.025%

    No Known Activations