Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 21136
    Prev
    Next
    INDEX
    Explanations

    ility

    np_max-act · gemini-2.0-flash

    This neuron activates on occurrences of the word “humility” (and its variants), i.e. it spots mentions of being humble.

    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
    -league
    -0.07
    >-->↵
    -0.06
    	ADD
    -0.06
    、彼
    -0.06
    Senior
    -0.06
     وج
    -0.06
     caches
    -0.06
     حتى
    -0.06
    *L
    -0.06
    bian
    -0.06
    POSITIVE LOGITS
    곡
    0.08
     ConfigureServices
    0.07
     Petr
    0.06
    �
    0.06
    tvrt
    0.06
    --){↵
    0.06
    ịnh
    0.06
    _CHECK
    0.06
     cryptoc
    0.06
    (body
    0.06
    Activations Density 0.016%

    No Known Activations