© 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. 2-RES-JB
    11. 12506
    Prev
    Next
    INDEX
    Explanations

    numbers represented in a different character encoding system

    oai_token-act-pair · gpt-3.5-turbo

    instances of characters or elements that may represent specific names or titles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    antha
    -0.95
    oran
    -0.94
    agar
    -0.90
    igree
    -0.84
    rators
    -0.80
    rified
    -0.80
    rim
    -0.80
    sonian
    -0.79
    ultz
    -0.79
    ittal
    -0.79
    POSITIVE LOGITS
    terday
    0.88
     annexed
    0.78
     princip
    0.77
    efe
    0.71
     compr
    0.70
     Hert
    0.69
     versa
    0.67
     incre
    0.66
    ierre
    0.65
     annex
    0.64
    Activations Density 0.016%

    No Known Activations