© 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. 26-LLAMASCOPE-RES-32K
    11. 5598
    Prev
    Next
    INDEX
    Explanations

    verb phrases indicating a process or action related to food preparation and consumption

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.26.hook_resid_post
    Hook Layer
    26
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    .MixedReality
    -0.16
    ÑģÑĤа
    -0.15
    cba
    -0.15
    legg
    -0.15
     gathering
    -0.15
     receive
    -0.14
    isbury
    -0.14
     Obtain
    -0.13
    кÑĢа
    -0.13
    ÑĪÑĮ
    -0.13
    POSITIVE LOGITS
     accompany
    0.32
     accompanies
    0.31
     complement
    0.29
     boot
    0.28
     tide
    0.27
    -boot
    0.25
     Tide
    0.24
    boot
    0.22
    _boot
    0.22
     accompagn
    0.22
    Activations Density 0.137%

    No Known Activations