© 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. 90810
    Prev
    Next
    INDEX
    Explanations

    Quantifiers (much, many)

    np_max-act · gemini-2.0-flash

    The neuron specifically fires on the concessive quantifier phrase “how much you,” as in “no matter how much you…”

    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
     Selbst
    -0.06
    (Data
    -0.06
     wrong
    -0.06
    الف
    -0.06
     emerge
    -0.06
     Yellow
    -0.06
     Wig
    -0.06
     relentlessly
    -0.06
     wreckage
    -0.06
    (UnmanagedType
    -0.06
    POSITIVE LOGITS
    /admin
    0.07
    tring
    0.07
    QUIRES
    0.07
    (strpos
    0.07
    .textAlignment
    0.07
    mî
    0.06
     zł
    0.06
    WEEN
    0.06
    	LL
    0.06
    DataService
    0.06
    Activations Density 0.015%

    No Known Activations