© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 12711
    Prev
    Next
    INDEX
    Explanations

    color

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of “color” (and its derivatives, e.g. “Color,” “colorimetric,” etc.) in the text.

    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
    申
    -0.07
    .jpa
    -0.07
    ssh
    -0.07
    .transaction
    -0.07
     ilişkin
    -0.07
    qtt
    -0.06
    SQLException
    -0.06
     Sabbath
    -0.06
     JA
    -0.06
    astle
    -0.06
    POSITIVE LOGITS
     color
    0.17
     Color
    0.17
    Color
    0.14
     colour
    0.13
    color
    0.12
     colors
    0.12
    Colors
    0.12
     Colour
    0.12
    	color
    0.11
    [color
    0.11
    Activations Density 0.050%

    No Known Activations