© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 44351
    Prev
    Next
    INDEX
    Explanations

    grip

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of gripping or grasping actions, especially the word “grip” (and its variants).

    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
     Hal
    -0.08
    卖
    -0.07
     Nova
    -0.07
     ashamed
    -0.07
     allo
    -0.06
     van
    -0.06
    _／
    -0.06
    26
    -0.06
     inval
    -0.06
    ￣￣￣
    -0.06
    POSITIVE LOGITS
     Grip
    0.13
     grip
    0.13
     gripping
    0.12
     grips
    0.11
     gri
    0.08
     dpi
    0.08
     GK
    0.07
     psycopg
    0.07
     Gri
    0.07
    	gbc
    0.07
    Activations Density 0.004%

    No Known Activations