© 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. 10380
    Prev
    Next
    INDEX
    Explanations

    advanced technology and features

    oai_token-act-pair · gpt-3.5-turbo

    mentions of advanced technologies or concepts

    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
    que
    -0.81
    obyl
    -0.79
    avering
    -0.77
    bah
    -0.75
    uku
    -0.74
    SIZE
    -0.73
    ihil
    -0.70
    phis
    -0.70
    uff
    -0.70
    ãģ®éŃĶ
    -0.70
    POSITIVE LOGITS
    settings
    0.87
    enhagen
    0.78
     mathematics
    0.76
     advanced
    0.74
     detection
    0.72
     advancement
    0.72
     versions
    0.71
     degrees
    0.71
     civilizations
    0.69
     technology
    0.68
    Activations Density 0.012%

    No Known Activations