© 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. 6886
    Prev
    Next
    INDEX
    Explanations

    sentiment analysis/emotion

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on mentions and explanations of “Sentiment analysis,” i.e. tokens in the text that describe that task.

    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
     muff
    -0.07
    vette
    -0.07
     scale
    -0.06
    $b
    -0.06
    Kal
    -0.06
     út
    -0.06
    imos
    -0.06
    けて
    -0.06
     scrub
    -0.06
     trunc
    -0.06
    POSITIVE LOGITS
     cloak
    0.06
     alteration
    0.06
     GRID
    0.06
    .AnchorStyles
    0.06
    .Grid
    0.06
    태
    0.06
     changing
    0.06
     elim
    0.06
     triệu
    0.06
     letterSpacing
    0.06
    Activations Density 0.008%

    No Known Activations