© 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. 17214
    Prev
    Next
    INDEX
    Explanations

    key phrases and concepts related to what truly matters or is significant in a discussion

    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
    ácil
    -0.15
    wish
    -0.14
     fab
    -0.14
    ÃŃc
    -0.14
    صÙģ
    -0.14
    ium
    -0.14
    aire
    -0.14
    OWN
    -0.14
    bor
    -0.14
    insky
    -0.14
    POSITIVE LOGITS
     matters
    0.87
     matter
    0.85
     mattered
    0.76
    matter
    0.76
     Matters
    0.74
     Matter
    0.73
     counts
    0.52
     Counts
    0.47
     count
    0.46
     counted
    0.43
    Activations Density 0.207%

    No Known Activations