© 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. 24-LLAMASCOPE-RES-32K
    11. 18966
    Prev
    Next
    INDEX
    Explanations

    negations and phrases related to limitations or contradictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    olia
    -0.14
    asley
    -0.13
    apus
    -0.13
    arger
    -0.13
    ountains
    -0.13
     còn
    -0.13
    åĩ¡
    -0.13
    )(((
    -0.13
    MMdd
    -0.13
    анÑģи
    -0.13
    POSITIVE LOGITS
     either
    1.62
    either
    1.38
     Either
    1.35
     EITHER
    1.29
    Either
    1.26
     либо
    0.77
     soit
    0.68
    ither
    0.68
     neither
    0.63
    ITHER
    0.60
    Activations Density 0.411%

    No Known Activations