© 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. 10416
    Prev
    Next
    INDEX
    Explanations

    titles or forms of address starting with "Sir."

    oai_token-act-pair · gpt-3.5-turbo

    titles or designations that include the word "Sir"

    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
    pace
    -0.70
    defense
    -0.66
     Pipeline
    -0.65
     kindergarten
    -0.65
    zona
    -0.65
     wrapper
    -0.64
    enders
    -0.63
    olina
    -0.63
    obos
    -0.62
    caliber
    -0.62
    POSITIVE LOGITS
     Sir
    3.69
    Sir
    2.96
     sir
    1.65
     Dame
    1.45
     Lord
    1.29
     Sheikh
    1.24
    Lord
    1.13
     Roger
    1.12
     Nigel
    1.07
     Trafford
    1.05
    Activations Density 0.012%

    No Known Activations