© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 28-LLAMASCOPE-RES-32K
    11. 23938
    Prev
    Next
    INDEX
    Explanations

    references to past actions and states, particularly using the verb "was"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ilt
    -0.17
    apl
    -0.16
    uds
    -0.16
    uda
    -0.15
    andes
    -0.15
    à¸ı
    -0.15
    enin
    -0.14
    Toolkit
    -0.14
    antlr
    -0.14
    ObjectContext
    -0.14
    POSITIVE LOGITS
     preceded
    0.31
    preced
    0.27
     preced
    0.23
     proceeded
    0.22
     proceed
    0.22
     employed
    0.20
     Proceed
    0.19
     survived
    0.18
     proceeding
    0.18
     prec
    0.17
    Activations Density 0.027%

    No Known Activations