© 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. 12601
    Prev
    Next
    INDEX
    Explanations

    answer

    np_max-act · gemini-2.0-flash

    The neuron activates on words indicating provided solutions or answer keys (e.g. “answer,” “solutions,” “key”).

    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
    Comfort
    -0.07
     beads
    -0.07
    asics
    -0.06
    Tv
    -0.06
     comfort
    -0.06
    ũng
    -0.06
    reveal
    -0.06
    _through
    -0.06
    орд
    -0.06
     요
    -0.06
    POSITIVE LOGITS
     exchanging
    0.07
     اقتصادی
    0.07
     technical
    0.06
    .Tasks
    0.06
    >';↵
    0.06
     Israel
    0.06
    Grupo
    0.06
     collective
    0.06
     предполаг
    0.06
    _SAMPLE
    0.06
    Activations Density 0.023%

    No Known Activations