© 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. 11007
    Prev
    Next
    INDEX
    Explanations

    references to finality or concluding elements, especially in the context of events or works

    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
    pery
    -0.18
    ed
    -0.17
    Ø®ÙĪاÙĨ
    -0.15
    edly
    -0.15
    esse
    -0.15
    annes
    -0.15
    eso
    -0.15
    eses
    -0.14
    een
    -0.14
    &W
    -0.14
    POSITIVE LOGITS
    ised
    0.33
    izing
    0.31
    ization
    0.29
    ity
    0.29
    mente
    0.29
    izers
    0.28
     touches
    0.28
    cial
    0.27
    izes
    0.27
    ists
    0.27
    Activations Density 0.042%

    No Known Activations