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. 26-LLAMASCOPE-RES-32K
    6. 6630
    Prev
    Next
    INDEX
    Explanations

    references to migration and international experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.26.hook_resid_post
    Hook Layer
    26
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    etto
    -0.17
    jong
    -0.16
     statewide
    -0.15
    atron
    -0.14
    kov
    -0.14
    ipeg
    -0.13
    _allocator
    -0.13
    lator
    -0.13
    Mappings
    -0.13
    BY
    -0.13
    POSITIVE LOGITS
     abroad
    0.72
     Ab
    0.68
     overseas
    0.57
    Ab
    0.54
     foreign
    0.53
     Overse
    0.52
    -ab
    0.47
    foreign
    0.46
    _ab
    0.44
    ab
    0.44
    Activations Density 0.207%

    No Known Activations