© 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. 126384
    Prev
    Next
    INDEX
    Explanations

    comfortable

    np_max-act · gemini-2.0-flash

    This neuron detects words that describe comfort or a cozy, relaxing environment.

    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
    inati
    -0.08
    Seed
    -0.08
    AI
    -0.07
    Div
    -0.07
     License
    -0.07
    _tuples
    -0.07
     gen
    -0.07
     seed
    -0.07
     læ
    -0.07
     aerial
    -0.06
    POSITIVE LOGITS
     comfortable
    0.15
     comfort
    0.13
     Comfort
    0.12
    Comfort
    0.12
     comfy
    0.10
     comfortably
    0.09
    comfort
    0.09
     confort
    0.09
    0.08
     uncomfortable
    0.08
    Activations Density 0.014%

    No Known Activations