© 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 (Instruct)
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 27055
    Prev
    Next
    INDEX
    Explanations

    self-references

    np_max-act · gemini-2.0-flash

    This neuron detects first-person self-references (first-person singular pronouns and equivalents like "I", "me", "myself", "Eu", etc.).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    semblies
    -0.07
    .optional
    -0.07
    .Tween
    -0.07
     Shield
    -0.07
     described
    -0.06
    gravity
    -0.06
     abras
    -0.06
     sockaddr
    -0.06
     信
    -0.06
     دش
    -0.06
    POSITIVE LOGITS
    角色
    0.07
    _given
    0.07
    ILL
    0.06
     idea
    0.06
     Tahoe
    0.06
    стич
    0.06
    ραν
    0.06
     nikdy
    0.06
    ��
    0.06
    _Metadata
    0.06
    Activations Density 0.182%

    No Known Activations