© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 12101
    Prev
    Next
    INDEX
    Explanations

    words related to architectural structures or locations

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific locations or entities related to redevelopment projects

    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
     Mafia
    -0.69
     caric
    -0.68
     Sonic
    -0.67
     Lip
    -0.65
    capt
    -0.64
     Silicon
    -0.62
     acceler
    -0.61
    icter
    -0.60
     Elijah
    -0.60
     Jess
    -0.60
    POSITIVE LOGITS
    unda
    3.42
    undo
    3.13
     Mund
    2.24
    und
    2.06
    ondo
    1.82
    undy
    1.30
     mund
    1.28
    mund
    1.26
     PP
    1.20
    redo
    1.16
    Activations Density 0.033%

    No Known Activations