© 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. 25-LLAMASCOPE-RES-32K
    11. 18426
    Prev
    Next
    INDEX
    Explanations

    words related to governance and management practices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
                    ↵
    -0.23
    story
    -0.19
    /student
    -0.19
     storm
    -0.19
     stor
    -0.19
     strokes
    -0.18
    _storage
    -0.18
    				↵
    -0.18
     stationed
    -0.17
    strict
    -0.17
    POSITIVE LOGITS
    bucks
    0.20
    cipher
    0.19
    coach
    0.19
    vation
    0.18
     quo
    0.17
    -alone
    0.17
    ois
    0.17
    (ST
    0.16
    pile
    0.16
    /testify
    0.16
    Activations Density 0.567%

    No Known Activations