© 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. 4-RES-JB
    11. 15618
    Prev
    Next
    INDEX
    Explanations

    references to comparative or relational concepts between different entities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "respective" and its variations

    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
    cher
    -0.77
    missing
    -0.76
    bay
    -0.75
    VL
    -0.75
    arer
    -0.73
    ker
    -0.72
    fish
    -0.70
    ARCH
    -0.70
    trak
    -0.70
    bler
    -0.69
    POSITIVE LOGITS
     respective
    1.05
     halves
    0.89
     genders
    0.88
     oppos
    0.87
     administrations
    0.84
     fortunes
    0.84
     sides
    0.83
     merits
    0.82
     constituencies
    0.82
     selves
    0.82
    Activations Density 0.011%

    No Known Activations