© 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. 104851
    Prev
    Next
    INDEX
    Explanations

    generic text

    np_max-act · gemini-2.0-flash

    The neuron fires strongly on abstract‐noun and modifier tokens that signal informational or expository content—words like “information,” “knowledge,” “including,” “based,” etc.

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

    references to techniques or treatments related to improving physical appearance, particularly in dental and sexual health contexts.

    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
    SYNC
    -0.07
     Joined
    -0.07
    	AND
    -0.06
     chunk
    -0.06
    이나
    -0.06
    NIL
    -0.06
    	record
    -0.06
    lu
    -0.06
    .mode
    -0.06
    dob
    -0.06
    POSITIVE LOGITS
     constitutional
    0.08
     comenz
    0.07
     электр
    0.06
     تض
    0.06
     plight
    0.06
    .parseDouble
    0.06
    .style
    0.06
     destac
    0.06
     gute
    0.06
    lbrace
    0.06
    Activations Density 0.458%

    No Known Activations