© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 7-RES-JB
    11. 5166
    Prev
    Next
    INDEX
    Explanations

    phrases related to collaboration, teamwork, and unity

    oai_token-act-pair · gpt-3.5-turbo

    themes related to social dynamics and relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     predecessor
    -0.64
     Investor
    -0.58
    é¾įå¥ij士
    -0.57
     successor
    -0.55
     entrepreneur
    -0.54
     advertis
    -0.53
     reader
    -0.53
     Seller
    -0.52
     restoration
    -0.52
    ainment
    -0.51
    POSITIVE LOGITS
     themselves
    1.17
    selves
    0.99
     theirs
    0.89
     alike
    0.87
    their
    0.85
     respectively
    0.81
     converge
    0.73
     THEIR
    0.73
     riot
    0.71
     perished
    0.71
    Activations Density 1.081%

    No Known Activations