© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 110176
    Prev
    Next
    INDEX
    Explanations

    math expressions

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens in the model’s own multi‐step explanatory reasoning (i.e. assistant-generated solution text) rather than on user queries.

    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
    Ja
    -0.08
     Hab
    -0.07
    stk
    -0.07
    Hora
    -0.06
     WR
    -0.06
     cf
    -0.06
    munition
    -0.06
     Selbst
    -0.06
    DIRECTORY
    -0.06
     bru
    -0.06
    POSITIVE LOGITS
     complying
    0.06
    ρθρο
    0.06
     élect
    0.06
    ®
    0.06
     />";↵
    0.06
     ".↵
    0.06
     doanh
    0.06
     yaygın
    0.06
    _BOOL
    0.06
    RAND
    0.06
    Activations Density 0.081%

    No Known Activations