© 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. 1-RES-JB
    11. 22674
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific names, especially "Vikas Swarup" and medical terms like "pancreatic cancer" and "pancreatitis"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific names, particularly "Vik" and "pancreatic" along with related terms to health and events in sports

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    endi
    -0.76
     Lomb
    -0.70
    NV
    -0.70
    wolf
    -0.66
    SOURCE
    -0.65
    guard
    -0.64
    wall
    -0.64
    ãģ®éŃĶ
    -0.64
    Vote
    -0.63
     mammal
    -0.62
    POSITIVE LOGITS
    aeus
    0.83
    icians
    0.83
    mingham
    0.77
     htt
    0.73
     externalToEVAOnly
    0.72
    hesis
    0.71
    akespe
    0.71
    NetMessage
    0.70
    merga
    0.70
    zsche
    0.70
    Activations Density 0.025%

    No Known Activations