© 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. 128781
    Prev
    Next
    INDEX
    Explanations

    calculate

    np_max-act · gemini-2.0-flash

    recursive functions related to mathematical sequences such as Fibonacci or factorial.

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

    The neuron fires on verbs that describe computing or calculating (e.g. “calculate,” “calculates,” “compute,” “computes”) especially in the context of defining a Fibonacci‐type function.

    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
     handling
    -0.08
     Jasmine
    -0.07
     decimals
    -0.07
    افي
    -0.07
    MOV
    -0.06
    _CLIP
    -0.06
     سور
    -0.06
    瑞
    -0.06
    طبي
    -0.06
    Esp
    -0.06
    POSITIVE LOGITS
    镇
    0.07
    학년도
    0.07
     سپتامبر
    0.06
     olduğ
    0.06
     inund
    0.06
     Modelo
    0.06
    _links
    0.06
    =args
    0.06
     nejsou
    0.06
     contaminants
    0.06
    Activations Density 0.033%

    No Known Activations