© 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. 109429
    Prev
    Next
    INDEX
    Explanations

    URLs

    np_max-act · gemini-2.0-flash

    The neuron responds to short, usually uppercase‐and‐digit sequences—i.e. the little alphanumeric chunks you see in URL query strings or other encoded parameter tokens.

    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
    	counter
    -0.08
     himself
    -0.07
     survive
    -0.06
    _qty
    -0.06
    опас
    -0.06
     windshield
    -0.06
     starter
    -0.06
    JsonValue
    -0.06
     AH
    -0.06
     domin
    -0.06
    POSITIVE LOGITS
    parameter
    0.06
     �
    0.06
     NEVER
    0.06
     textStatus
    0.06
    oại
    0.06
     silicone
    0.06
    訓
    0.06
    블
    0.06
    .TextAlign
    0.06
    ，与
    0.06
    Activations Density 0.004%

    No Known Activations