Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 120
    Prev
    Next
    INDEX
    Explanations

    phrases related to acronyms like "DE" and "DEF"

    oai_token-act-pair · gpt-3.5-turbo

    references to the abbreviation "DE" and its variations in the context of definitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     peac
    -0.70
     betting
    -0.65
    ihad
    -0.65
     Yar
    -0.64
     unthinkable
    -0.61
     gust
    -0.58
     Pacers
    -0.58
     warm
    -0.58
     lanes
    -0.57
     bulb
    -0.57
    POSITIVE LOGITS
    DE
    4.12
    DEF
    1.79
     DE
    1.72
    DES
    1.57
    DEP
    1.57
    CE
    1.48
    de
    1.47
    DR
    1.45
    DA
    1.43
    BE
    1.42
    Activations Density 0.011%

    No Known Activations