© 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. 17349
    Prev
    Next
    INDEX
    Explanations

    phrases related to participation

    oai_token-act-pair · gpt-3.5-turbo

    references to participation in activities or processes

    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
     brittle
    -0.75
    ndra
    -0.69
     indebted
    -0.66
     spiders
    -0.64
     clocks
    -0.64
    worms
    -0.62
     engineers
    -0.62
    device
    -0.61
     suited
    -0.61
     hackers
    -0.60
    POSITIVE LOGITS
    ordinate
    1.06
    animate
    0.92
     lieu
    0.81
    cheon
    0.80
     unison
    0.76
     conversation
    0.76
     vitro
    0.74
     vivo
    0.72
     accordance
    0.72
     behalf
    0.72
    Activations Density 0.117%

    No Known Activations