© 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. 8903
    Prev
    Next
    INDEX
    Explanations

    Cole

    np_max-act · gemini-2.0-flash

    The neuron never reliably activates on any token and thus doesn’t detect any meaningful pattern.

    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
     Glow
    -0.06
     تهران
    -0.06
     Fucking
    -0.06
     Vậy
    -0.06
     Jo
    -0.06
    ities
    -0.06
    QualifiedName
    -0.06
     Fuck
    -0.05
     Sean
    -0.05
     Soph
    -0.05
    POSITIVE LOGITS
    AFE
    0.08
    crud
    0.07
    wig
    0.07
    (secret
    0.07
    \.
    0.07
     ↵
    0.07
    arians
    0.07
                               ↵
    0.07
     whenever
    0.07
    erv
    0.07
    Activations Density 0.030%

    No Known Activations