© 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. 26-LLAMASCOPE-RES-32K
    11. 20959
    Prev
    Next
    INDEX
    Explanations

    elements and processes related to planning and building structures or systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.26.hook_resid_post
    Hook Layer
    26
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     necessarily
    -0.17
     yoksa
    -0.16
    öl
    -0.16
    uji
    -0.15
    erer
    -0.13
    /if
    -0.13
     somewhere
    -0.13
    ãģ§ãģ¯ãģªãģĦ
    -0.13
    uran
    -0.12
     somehow
    -0.12
    POSITIVE LOGITS
     almost
    0.77
     virtually
    0.66
    almost
    0.66
     nearly
    0.63
     Almost
    0.59
    Almost
    0.58
    åĩłä¹İ
    0.52
     pretty
    0.52
     any
    0.52
     ANY
    0.50
    Activations Density 0.423%

    No Known Activations