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. 21021
    Prev
    Next
    INDEX
    Explanations

    phrases related to activities or concepts coming to completion or encompassing a period of time

    oai_token-act-pair · gpt-3.5-turbo

    words related to being bound or restricted

    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
     hacks
    -0.65
    zie
    -0.61
     schemes
    -0.61
     Sun
    -0.61
     devil
    -0.60
     requests
    -0.59
     whis
    -0.58
    Sche
    -0.58
     Powers
    -0.58
     Nero
    -0.57
    POSITIVE LOGITS
    OUND
    3.88
    ounded
    2.33
    amous
    1.94
    ounding
    1.87
    ound
    1.49
    OUN
    1.41
    ounds
    1.33
    ersion
    0.93
    Interstitial
    0.91
    ablished
    0.90
    Activations Density 0.013%

    No Known Activations