© 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. 31-LLAMASCOPE-RES-32K
    11. 2296
    Prev
    Next
    INDEX
    Explanations

    references to specific locations and experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     arz
    -0.16
    istrovstvÃŃ
    -0.14
     Azerbai
    -0.14
     Calgary
    -0.14
    erez
    -0.14
     Farage
    -0.14
    vette
    -0.14
    ktion
    -0.13
     decode
    -0.13
    mares
    -0.13
    POSITIVE LOGITS
     Wald
    0.44
     Concord
    0.41
     Emerson
    0.40
     Th
    0.39
     WAL
    0.32
     Henry
    0.31
     transcend
    0.30
    wald
    0.29
     Hawth
    0.28
     Ralph
    0.28
    Activations Density 0.021%

    No Known Activations