© 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. 126624
    Prev
    Next
    INDEX
    Explanations

    research findings

    np_max-act · gemini-2.0-flash

    This neuron detects language related to the evaluation and refinement phase of a process (e.g., phrases like “Based on the evaluation…” and “if… then…”).

    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
    ANDROID
    -0.07
     Shared
    -0.06
    .bindingNavigator
    -0.06
    Fetch
    -0.06
    Под
    -0.06
    اصر
    -0.06
     indispensable
    -0.06
     recur
    -0.05
    .Cmd
    -0.05
    .ย
    -0.05
    POSITIVE LOGITS
    997
    0.07
    ¨¨
    0.06
     ''↵↵
    0.06
    801
    0.06
     кни
    0.06
     vyrá
    0.06
    VG
    0.06
    _um
    0.06
    세요
    0.06
    CREASE
    0.06
    Activations Density 0.022%

    No Known Activations