© 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. 41478
    Prev
    Next
    INDEX
    Explanations

    Predictions about the future

    np_max-act · gemini-2.0-flash

    The neuron fires on small auxiliary or hedge words (like “would,” “will,” “just,” “can”) that signal possibility, intention, or conditional phrasing.

    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
    Reduce
    -0.07
    .registration
    -0.07
    MH
    -0.07
    abella
    -0.07
    อาช
    -0.07
    ;;
    -0.06
    WithEmailAndPassword
    -0.06
     Meghan
    -0.06
    iyon
    -0.06
    assistant
    -0.06
    POSITIVE LOGITS
    .Relative
    0.07
    odings
    0.06
     sequences
    0.06
     viewers
    0.06
    .X
    0.06
     StObject
    0.06
    ****/↵
    0.06
     pissed
    0.06
     Yuan
    0.06
    _markers
    0.06
    Activations Density 0.051%

    No Known Activations