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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. Residual Stream
    5. 24-LLAMASCOPE-RES-32K
    6. 18017
    Prev
    Next
    INDEX
    Explanations

    phrases indicating ongoing processes or states and their implications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    मत
    -0.15
    odb
    -0.15
    ClientRect
    -0.13
    ç¿°
    -0.12
    _HDR
    -0.12
    \Middleware
    -0.12
    Ïģιά
    -0.12
    ovid
    -0.12
    nahme
    -0.12
    wyn
    -0.12
    POSITIVE LOGITS
     still
    1.39
    still
    1.27
     Still
    1.21
    Still
    1.17
     STILL
    1.15
    ä»į
    0.91
     ainda
    0.85
     ancora
    0.81
     masih
    0.81
     vẫn
    0.79
    Activations Density 1.392%

    No Known Activations