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. 20826
    Prev
    Next
    INDEX
    Explanations

    pronunciations of various words or phrases

    oai_token-act-pair · gpt-3.5-turbo

    references to pronunciation

    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
     looted
    -0.78
     Committees
    -0.76
     returned
    -0.69
     recovered
    -0.65
     Saving
    -0.64
     Lib
    -0.62
     recover
    -0.62
     Images
    -0.61
     restored
    -0.61
     collections
    -0.60
    POSITIVE LOGITS
    pron
    5.00
     pron
    1.58
    plet
    1.01
    sembly
    1.01
    pter
    0.99
    ãĥ¼ãĥĨãĤ£
    0.98
    ãĥĩãĤ£
    0.96
    ËĪ
    0.94
     pronounced
    0.94
    ãĤ®
    0.92
    Activations Density 0.013%

    No Known Activations