© 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. 4-RES-JB
    6. 9539
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal or criminal actions, specifically involving documentation and events like shootings, licenses, and security measures

    oai_token-act-pair · gpt-3.5-turbo

    punctuation, specifically full stops and periods

    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
     administrator
    -0.78
     bunny
    -0.75
     imperson
    -0.74
     posture
    -0.74
     ribbon
    -0.73
     shield
    -0.73
     portrait
    -0.72
     advis
    -0.72
     herd
    -0.71
     sway
    -0.71
    POSITIVE LOGITS
     However
    1.22
     Additionally
    1.18
     Interestingly
    1.14
     Afterwards
    1.14
     Similarly
    1.12
     Luckily
    1.11
    sbm
    1.09
     Unfortunately
    1.09
     Meanwhile
    1.08
     Fortunately
    1.06
    Activations Density 0.558%

    No Known Activations