© 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. 25-LLAMASCOPE-RES-32K
    11. 22448
    Prev
    Next
    INDEX
    Explanations

    descriptors related to product features and benefits

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iaux
    -0.15
    ARAM
    -0.15
    iali
    -0.15
     Bookmark
    -0.14
    TypeInfo
    -0.13
     Shel
    -0.13
    phem
    -0.13
     Gregg
    -0.13
    IFn
    -0.13
     embedded
    -0.13
    POSITIVE LOGITS
     package
    0.63
     kit
    0.54
     Package
    0.54
     packages
    0.53
     bundle
    0.51
    -package
    0.48
     Kit
    0.46
    package
    0.46
     pack
    0.46
     kits
    0.45
    Activations Density 0.182%

    No Known Activations