© 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. 125948
    Prev
    Next
    INDEX
    Explanations

    ASCII characters

    np_max-act · gemini-2.0-flash

    ASCII art representations of shapes, particularly circles.

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

    This neuron activates on mentions of ASCII art—especially the tokens “ASCII” and “art” when the user is asking for or describing ASCII drawings.

    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
    /navigation
    -0.08
     AppRoutingModule
    -0.06
    	bs
    -0.06
     melt
    -0.06
    alculate
    -0.06
    .presenter
    -0.06
     bass
    -0.06
     magg
    -0.06
    PictureBox
    -0.06
     ^{↵
    -0.06
    POSITIVE LOGITS
     sne
    0.07
    .stub
    0.06
     matters
    0.06
    Fal
    0.06
     rég
    0.06
    RegExp
    0.06
    ійного
    0.06
    (?:
    0.06
    making
    0.06
     expiry
    0.06
    Activations Density 0.010%

    No Known Activations