© 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. 87531
    Prev
    Next
    INDEX
    Explanations

    needs and wants

    np_max-act · gemini-2.0-flash

    sentences or phrases in a formal/informational tone that state policies, commitments, instructions, or obligations (e.g., policy language and reassurance/attention phrases).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31

    needs and concerns

    np_max-act · gpt-5-miniTriggered by @yooniel31
    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
    Sex
    -0.07
    uro
    -0.06
    alloween
    -0.06
     singer
    -0.06
    edriver
    -0.06
     $
    -0.06
     –
    -0.06
    ungeon
    -0.06
     teachers
    -0.06
    เซ
    -0.06
    POSITIVE LOGITS
    TexParameter
    0.07
     tecr
    0.07
    無しさん
    0.07
            
↵        
↵
    0.06
     стал
    0.06
    ніцип
    0.06
    _fact
    0.06
    �
    0.06
     资
    0.06
     Chúa
    0.06
    Activations Density 0.112%

    No Known Activations