© 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. 6-LLAMASCOPE-RES-32K
    11. 7489
    Prev
    Next
    INDEX
    Explanations

    terms related to leaks or leakage

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.6.hook_resid_post
    Hook Layer
    6
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ignon
    -0.15
    амеÑĤ
    -0.15
    бÑĥ
    -0.14
    azzi
    -0.14
    AIT
    -0.14
    illet
    -0.14
    меÑĤ
    -0.14
    flies
    -0.14
    ilon
    -0.14
    gan
    -0.14
    POSITIVE LOGITS
    adier
    0.17
    ler
    0.17
    iera
    0.15
    ĴĪ
    0.15
    érica
    0.15
    ropolis
    0.14
    еÑĢп
    0.14
    stddef
    0.14
    edImage
    0.14
    .Stream
    0.14
    Activations Density 0.017%

    No Known Activations