© 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. 4-RES-JB
    11. 10379
    Prev
    Next
    INDEX
    Explanations

    specific references to the term "Liberator" or variations of it

    oai_token-act-pair · gpt-3.5-turbo

    variations of the word "Liber" or related forms, indicating a focus on concepts associated with liberation or freedom

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

    No Comments

    Negative Logits
    enegger
    -0.77
     à¨
    -0.76
    soDeliveryDate
    -0.74
    WARE
    -0.72
    WAY
    -0.66
     Cth
    -0.65
    IELD
    -0.64
     wig
    -0.63
    BOOK
    -0.61
     deaf
    -0.60
    POSITIVE LOGITS
    tarians
    1.42
    tarian
    1.05
    itas
    1.02
    ated
    1.00
    ator
    1.00
    ations
    0.95
    itary
    0.92
    iously
    0.92
    ators
    0.92
    aine
    0.91
    Activations Density 0.006%

    No Known Activations