© 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. 28-LLAMASCOPE-RES-32K
    11. 30132
    Prev
    Next
    INDEX
    Explanations

    phrases related to gaming consoles and industry discussions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Reddit
    -0.19
     subreddit
    -0.18
     stream
    -0.18
     reddit
    -0.18
    https
    -0.18
     selfies
    -0.17
    Stream
    -0.17
     Stream
    -0.17
    "https
    -0.17
     Reddit
    -0.17
    POSITIVE LOGITS
     PSP
    0.25
     DS
    0.22
     Nam
    0.20
     SCE
    0.19
     ds
    0.19
     Medal
    0.19
     Kinect
    0.18
     Wii
    0.18
     Lips
    0.18
     downloadable
    0.17
    Activations Density 0.156%

    No Known Activations