© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 5511
    Prev
    Next
    INDEX
    Explanations

    mentions of time periods denoted by "late" followed by a year or decade

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "late" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    roma
    -0.89
    ebook
    -0.82
    \\\\\\\\
    -0.79
    adr
    -0.77
    tre
    -0.76
    qua
    -0.75
    utenberg
    -0.74
    iframe
    -0.74
    ortion
    -0.73
    alach
    -0.72
    POSITIVE LOGITS
     afternoon
    1.06
     evening
    0.88
     late
    0.82
     nights
    0.81
     adolescence
    0.80
     antiquity
    0.80
     night
    0.79
     morning
    0.78
     evenings
    0.75
     lunch
    0.70
    Activations Density 0.017%

    No Known Activations