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. 31-LLAMASCOPE-RES-32K
    6. 18263
    Prev
    Next
    INDEX
    Explanations

    expressions related to human evolution and cooperation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éĨ
    -0.15
     Unblock
    -0.15
    IMER
    -0.15
    .currentThread
    -0.14
    ίδα
    -0.13
    ridge
    -0.13
    olor
    -0.13
     Unidos
    -0.13
     Registry
    -0.13
    .getServer
    -0.13
    POSITIVE LOGITS
     hunter
    0.37
     Hunter
    0.30
    hunter
    0.28
    Hunter
    0.26
     evolutionary
    0.25
     Pale
    0.24
     hunters
    0.24
     hunting
    0.24
     selection
    0.23
     ancestral
    0.22
    Activations Density 0.024%

    No Known Activations