Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 85077
    Prev
    Next
    INDEX
    Explanations

    forum conversations

    np_max-act · gemini-2.0-flash

    The neuron primarily detects occurrences of the special `<|start_header_id|>` token.

    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
     motivate
    -0.07
    opaque
    -0.07
    arching
    -0.06
    �
    -0.06
    )]↵
    -0.06
    .IsEnabled
    -0.06
    .non
    -0.06
     θα
    -0.06
    _RETURNTRANSFER
    -0.06
     Unless
    -0.06
    POSITIVE LOGITS
     Running
    0.07
     Fathers
    0.07
    /command
    0.07
    =q
    0.07
    _sep
    0.07
    Runtime
    0.06
     playlist
    0.06
     presidents
    0.06
     fathers
    0.06
     totals
    0.06
    Activations Density 0.017%

    No Known Activations