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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 30-LLAMASCOPE-RES-32K
    6. 17919
    Prev
    Next
    INDEX
    Explanations

    code operations related to messaging and event creation in a programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Moor
    -0.16
     Maul
    -0.15
     Marilyn
    -0.15
     Mach
    -0.15
    Mob
    -0.14
     Monk
    -0.13
    ÙħÙĪÙĦ
    -0.13
     Mobility
    -0.13
    mute
    -0.13
     Mold
    -0.13
    POSITIVE LOGITS
     message
    0.96
     Message
    0.83
     messages
    0.81
    message
    0.79
     mess
    0.78
    -message
    0.75
     msg
    0.74
    Message
    0.73
     Mess
    0.71
    _message
    0.71
    Activations Density 0.282%

    No Known Activations