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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 21-LLAMASCOPE-RES-32K
    6. 16498
    Prev
    Next
    INDEX
    Explanations

    specific temporal markers indicating stages, points, or phases in discussions or narratives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     at
    -0.24
    ãn
    -0.16
    ighth
    -0.15
    cased
    -0.14
    emoc
    -0.14
    ONUS
    -0.14
    è¡Ĺéģĵ
    -0.14
    änge
    -0.14
    udas
    -0.14
    rawler
    -0.14
    POSITIVE LOGITS
     level
    0.26
     expense
    0.25
     discretion
    0.25
     intervals
    0.24
     jun
    0.23
     glance
    0.21
     Expense
    0.20
     helm
    0.19
     site
    0.19
    expense
    0.18
    Activations Density 0.475%

    No Known Activations