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. 3188
    Prev
    Next
    INDEX
    Explanations

    keywords related to urgent or important matters

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate urgent or important actions

    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
    aa
    -0.69
    omer
    -0.63
    ety
    -0.61
    365
    -0.61
    ooters
    -0.59
    reference
    -0.59
     belong
    -0.58
     spree
    -0.58
     Sites
    -0.57
    christ
    -0.57
    POSITIVE LOGITS
     pressing
    3.96
     pressed
    1.83
     presses
    1.81
     press
    1.50
    pressed
    1.47
    press
    1.41
    Press
    1.33
     Press
    1.33
     pressuring
    1.30
     urgent
    1.29
    Activations Density 0.017%

    No Known Activations