© 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. 21-LLAMASCOPE-RES-32K
    11. 15717
    Prev
    Next
    INDEX
    Explanations

    financial metrics and projections related to investments and funding activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     decades
    -0.28
    30
    -0.26
     Thirty
    -0.25
    Thirty
    -0.23
     Fifty
    -0.23
    40
    -0.21
    25
    -0.20
    20
    -0.20
    50
    -0.20
     century
    -0.19
    POSITIVE LOGITS
    ä¸īå¹´
    0.44
     four
    0.39
     three
    0.36
     five
    0.35
     fours
    0.27
    four
    0.27
    five
    0.26
     THREE
    0.26
    three
    0.26
     two
    0.24
    Activations Density 0.155%

    No Known Activations