© 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. 11060
    Prev
    Next
    INDEX
    Explanations

    statistical measures and percentages

    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
    æŁĦ
    -0.16
    undles
    -0.14
    uyá»ħn
    -0.14
     Millenn
    -0.14
     cog
    -0.14
    ordum
    -0.14
    éijij
    -0.14
    ewire
    -0.14
    eu
    -0.14
     tat
    -0.13
    POSITIVE LOGITS
    95
    0.49
     CI
    0.46
    CI
    0.45
     confidence
    0.43
     Confidence
    0.40
    confidence
    0.36
    -confidence
    0.33
    _CI
    0.33
     ci
    0.30
    _ci
    0.29
    Activations Density 0.008%

    No Known Activations