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. 5862
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "Beatitude."

    oai_token-act-pair · gpt-3.5-turbo

    references to "Beat" and "Beatles."

    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
    aneous
    -0.76
    reen
    -0.69
    tu
    -0.66
    ary
    -0.64
     cul
    -0.63
     caution
    -0.63
    ici
    -0.62
    tti
    -0.62
     linkage
    -0.61
    alg
    -0.61
    POSITIVE LOGITS
     Beat
    4.02
    Beat
    2.94
    beat
    2.27
     beat
    1.92
     Beats
    1.50
     beats
    1.42
     beaten
    1.28
     Beatles
    1.27
     beating
    1.24
     Blitz
    1.09
    Activations Density 0.011%

    No Known Activations