© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 4621
    Prev
    Next
    INDEX
    Explanations

    numbers and comparisons

    oai_token-act-pair · gpt-3.5-turbo

    qualifiers and phrases indicating emphasis or comparison

    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
     Gothic
    -0.67
     brut
    -0.62
     preparation
    -0.62
     Secondary
    -0.61
     Tsukuyomi
    -0.60
    istant
    -0.58
     hanging
    -0.58
     exiting
    -0.58
     weeping
    -0.57
     Cinderella
    -0.57
    POSITIVE LOGITS
    wrote
    1.19
    etheless
    1.10
    ifies
    1.10
    poses
    1.09
     resembled
    1.08
    iates
    1.05
     resembles
    1.04
    did
    1.03
     embodies
    1.00
    should
    1.00
    Activations Density 0.578%

    No Known Activations