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. 21945
    Prev
    Next
    INDEX
    Explanations

    the word "uh"

    oai_token-act-pair · gpt-3.5-turbo

    the expression of disbelief or surprise

    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
     provision
    -0.74
     cop
    -0.67
     sustained
    -0.66
     closure
    -0.65
     accessible
    -0.63
     compounded
    -0.63
     contained
    -0.63
     branded
    -0.62
     cottage
    -0.62
     Giants
    -0.62
    POSITIVE LOGITS
    uh
    4.05
    ugh
    1.51
    Uh
    1.37
    uf
    1.37
    oh
    1.36
     uh
    1.29
    uz
    1.29
    eh
    1.28
    ah
    1.28
    uy
    1.27
    Activations Density 0.012%

    No Known Activations