© 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. 18489
    Prev
    Next
    INDEX
    Explanations

    terms related to property damage or loss

    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
     Ye
    -0.17
     YE
    -0.16
    lox
    -0.15
     Yard
    -0.15
    fos
    -0.15
    äter
    -0.15
    änn
    -0.15
    ÄIJT
    -0.14
    igte
    -0.14
    Å¡ku
    -0.14
    POSITIVE LOGITS
    y
    1.28
    yb
    0.51
    yw
    0.50
    ythe
    0.46
    yh
    0.43
    yth
    0.41
    yk
    0.40
    yro
    0.38
    yg
    0.38
    yi
    0.38
    Activations Density 0.043%

    No Known Activations