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. 8-RES-JB
    6. 16395
    Prev
    Next
    INDEX
    Explanations

    phrases expressing disbelief or frustration

    oai_token-act-pair · gpt-3.5-turbo

    phrases and expressions emphasizing disbelief or frustration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    terday
    -0.77
    CTV
    -0.71
    uria
    -0.68
    icho
    -0.65
    orea
    -0.65
    968
    -0.63
    culosis
    -0.62
    icia
    -0.61
    provided
    -0.61
     indefinitely
    -0.60
    POSITIVE LOGITS
     heck
    1.54
     hell
    1.41
     fuck
    1.32
     Heck
    1.05
     difference
    0.94
     HELL
    0.93
     devil
    0.93
     Fuck
    0.91
     slightest
    0.86
     feds
    0.85
    Activations Density 0.099%

    No Known Activations