© 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. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 9-LLAMASCOPE-RES-32K
    11. 24038
    Prev
    Next
    INDEX
    Explanations

    pronouns related to direct address and references to the reader

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     themselves
    -0.18
     itself
    -0.18
    atti
    -0.16
    tering
    -0.14
    ly
    -0.14
    woord
    -0.14
    lÃŃ
    -0.14
    ाà¤Ĺत
    -0.14
     you
    -0.14
     usted
    -0.13
    POSITIVE LOGITS
    /us
    0.30
    -même
    0.24
    /her
    0.24
    zelf
    0.23
    SELF
    0.22
    åĢij
    0.19
    self
    0.18
    ’re
    0.17
     guys
    0.17
    itable
    0.15
    Activations Density 0.120%

    No Known Activations