© 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. 19-RESID-POST-AA
    11. 57800
    Prev
    Next
    INDEX
    Explanations

    ari

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_19/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    "Yes
    -0.07
    —for
    -0.07
     vlan
    -0.06
    “Yes
    -0.06
     เช
    -0.06
    ].↵↵
    -0.06
    snd
    -0.06
     гід
    -0.06
    }"↵
    -0.06
    cat
    -0.06
    POSITIVE LOGITS
    are
    0.08
    ara
    0.08
    라피
    0.07
    ABEL
    0.07
    ael
    0.07
    -entry
    0.07
    ari
    0.07
    ARE
    0.07
     acess
    0.07
    ARI
    0.07
    Activations Density 0.033%

    No Known Activations