Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. MLP
    5. 22-LLAMASCOPE-MLP-32K
    6. 32697
    Prev
    Next
    INDEX
    Explanations

    mentions of scientific studies and research findings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.22.hook_mlp_out
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    andy
    -0.15
     erken
    -0.14
    usercontent
    -0.13
    ÙħاÙĨÛĮ
    -0.13
    cheng
    -0.13
     mekt
    -0.12
     Amateur
    -0.12
    oric
    -0.12
    strup
    -0.12
    udo
    -0.12
    POSITIVE LOGITS
     examination
    0.45
     Exam
    0.41
     examine
    0.40
    exam
    0.39
     exam
    0.38
    Exam
    0.38
     examined
    0.38
     examining
    0.38
     Examination
    0.37
     examines
    0.36
    Activations Density 0.431%

    No Known Activations