© 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. 2-LLAMASCOPE-RES-32K
    11. 21993
    Prev
    Next
    INDEX
    Explanations

    words related to stationary objects and items

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ане
    -0.14
    agem
    -0.14
     spec
    -0.14
    ìłķìĿĦ
    -0.13
    nap
    -0.13
    ants
    -0.13
    иÑİ
    -0.13
    aram
    -0.13
    oday
    -0.13
     Meal
    -0.13
    POSITIVE LOGITS
    .jquery
    0.16
    urd
    0.15
    ardy
    0.15
    ibold
    0.14
    eward
    0.14
    tries
    0.14
    osti
    0.14
     Raz
    0.14
    atism
    0.14
    ibi
    0.14
    Activations Density 0.013%

    No Known Activations