© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 13104
    Prev
    Next
    INDEX
    Explanations

    names of characters and storylines in various forms of entertainment

    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
    adder
    -0.18
    vise
    -0.15
    #aa
    -0.15
    ardown
    -0.15
    eyle
    -0.15
    AdapterManager
    -0.15
    uits
    -0.15
    adele
    -0.15
     fellow
    -0.15
    eea
    -0.15
    POSITIVE LOGITS
     protagonist
    0.23
     protagonists
    0.21
     young
    0.21
     ambit
    0.19
     ordinary
    0.19
    主人
    0.17
     titular
    0.16
     main
    0.16
     protagon
    0.16
     lon
    0.15
    Activations Density 0.291%

    No Known Activations