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. 4-RES-JB
    6. 22830
    Prev
    Next
    INDEX
    Explanations

    words containing the sequence "ugh"

    oai_token-act-pair · gpt-3.5-turbo

    expressions of frustration or exasperation

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

    No Comments

    Negative Logits
    åĤ
    -0.68
    BIL
    -0.67
     Butterfly
    -0.65
     Takeru
    -0.63
    Guard
    -0.62
     Devils
    -0.61
     Sanct
    -0.60
     Defenders
    -0.60
    Level
    -0.60
     Luthor
    -0.59
    POSITIVE LOGITS
    ugh
    1.03
    ttp
    0.97
    awk
    0.90
    urst
    0.90
    uay
    0.85
    ricanes
    0.84
    ftime
    0.82
    ushi
    0.80
    azard
    0.79
    ricane
    0.78
    Activations Density 0.005%

    No Known Activations