© 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. 17-LLAMASCOPE-RES-32K
    11. 15382
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "after all" and other expressions that indicate reflection or reconsideration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    areth
    -0.17
    eniable
    -0.16
    aku
    -0.16
    serrat
    -0.15
    vens
    -0.15
    bbe
    -0.15
    baugh
    -0.14
    utt
    -0.14
     Chamber
    -0.14
    s
    -0.13
    POSITIVE LOGITS
    æ¯ķ
    0.27
     after
    0.27
     pÅĻece
    0.25
     AFTER
    0.23
    after
    0.22
     After
    0.22
    After
    0.21
    inde
    0.21
     totiž
    0.21
     indeed
    0.21
    Activations Density 0.093%

    No Known Activations