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. 4408
    Prev
    Next
    INDEX
    Explanations

    references to specific entities or concepts related to superstition

    oai_token-act-pair · gpt-3.5-turbo

    terms related to superstition and cellular technology

    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
    ibur
    -0.94
    ester
    -0.71
    ura
    -0.71
    ounce
    -0.70
    renheit
    -0.69
     vape
    -0.69
    ellen
    -0.68
     vows
    -0.67
    toe
    -0.67
    collar
    -0.66
    POSITIVE LOGITS
     superst
    1.78
     cellular
    1.55
     Pag
    1.31
     Fowler
    1.28
     pag
    1.16
     Anaheim
    1.16
     trunk
    1.11
     Din
    1.05
     Lansing
    0.99
     Yankees
    0.98
    Activations Density 0.033%

    No Known Activations