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

    Neuronpedia

    Jacobian LensNEW
    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. 11-RES-JB
    11. 22000
    Prev
    Next
    INDEX
    Explanations

    pronouns or humanoid subjects and their actions

    oai_token-act-pair · gpt-3.5-turbo

    references to actions taken by individuals or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aukee
    -0.68
    yip
    -0.65
    ãĤ´ãĥ³
    -0.64
    Skip
    -0.61
    Īè
    -0.58
     practition
    -0.57
     Moroc
    -0.57
     Skip
    -0.57
    Same
    -0.56
     Previous
    -0.56
    POSITIVE LOGITS
     accordingly
    1.24
     consequently
    1.20
     certainly
    1.12
     therefore
    1.06
     promptly
    0.94
     deserve
    0.93
     consequ
    0.92
     definitely
    0.92
     intend
    0.92
     frankly
    0.90
    Activations Density 0.271%

    No Known Activations