© 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. 11-RESID-POST-AA
    11. 25897
    Prev
    Next
    INDEX
    Explanations

    importance

    np_max-act · gemini-2.0-flash

    conversational exchanges and queries about learning or improving skills.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    It detects emphasis cues in instructional or advisory text that signal key or important points.

    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
    经过
    -0.07
     teşkil
    -0.07
     Moor
    -0.06
     bv
    -0.06
    шь
    -0.06
     budgets
    -0.06
     euler
    -0.06
    ेखत
    -0.06
     حسین
    -0.06
    _Valid
    -0.06
    POSITIVE LOGITS
    นาด
    0.07
     pick
    0.06
    w
    0.06
     affili
    0.06
     FW
    0.06
    ประ
    0.06
     pä
    0.06
     Prahy
    0.06
     reject
    0.06
     τρα
    0.06
    Activations Density 0.039%

    No Known Activations