© 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. 0-RES-JB
    11. 1424
    Prev
    Next
    INDEX
    Explanations

    words related to intruders or unauthorized access

    oai_token-act-pair · gpt-3.5-turbo

    terms related to intruders and specific locations, particularly referencing "Corsicana."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Telecom
    -0.76
     Pai
    -0.73
     Quantum
    -0.68
     Uk
    -0.65
    Nat
    -0.65
     dividend
    -0.63
    writing
    -0.62
     Bermuda
    -0.62
     stump
    -0.61
     Regulatory
    -0.61
    POSITIVE LOGITS
    uder
    4.04
    uders
    2.56
     Cors
    1.51
    kson
    1.15
    itton
    1.01
    asher
    0.98
    auder
    0.95
    igne
    0.95
    isle
    0.91
     Rouse
    0.90
    Activations Density 0.052%

    No Known Activations