© 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. 6986
    Prev
    Next
    INDEX
    Explanations

    Mistakes

    np_max-act · gemini-2.0-flash

    The neuron flags words that signal mistakes or accidental errors (e.g. “mistakenly,” “accidentally,” “wrong,” “brought” in the sense of a mix-up).

    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
    included
    -0.08
    .CreateDirectory
    -0.07
     Bris
    -0.07
     Reward
    -0.07
    _scaled
    -0.07
     dog
    -0.07
     adultery
    -0.06
     ListNode
    -0.06
     kosher
    -0.06
    Hen
    -0.06
    POSITIVE LOGITS
     forme
    0.07
     먹
    0.07
    requests
    0.06
     изображ
    0.06
     свои
    0.06
    uario
    0.06
    álo
    0.06
    }↵↵↵↵↵
    0.06
     HAPP
    0.06
     Yönetim
    0.06
    Activations Density 0.045%

    No Known Activations