© 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. 6358
    Prev
    Next
    INDEX
    Explanations

    letter

    np_max-act · gemini-2.0-flash

    This neuron activates on the word “letter,” specifically picking up instructions that say to choose a letter as the answer.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    logical conclusions based on premises and assumptions.

    oai_token-act-pair · gpt-4o-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
    ========
    -0.07
     Observable
    -0.07
    �
    -0.07
    PN
    -0.06
    اصل
    -0.06
    _CN
    -0.06
     sane
    -0.06
    ispens
    -0.06
    istické
    -0.06
     온라인
    -0.06
    POSITIVE LOGITS
    DAY
    0.07
     čtvrt
    0.06
     vapor
    0.06
     responder
    0.06
    leon
    0.06
     педагог
    0.06
    strt
    0.06
    _EDGE
    0.06
    ivic
    0.06
     hdr
    0.06
    Activations Density 0.002%

    No Known Activations