© 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. 2019
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to events or situations narrowly avoiding a negative outcome

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "narrowly" indicating close calls or narrowly avoided situations

    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
     Occupations
    -0.74
     Founding
    -0.69
    upon
    -0.66
     Apostles
    -0.65
     Ancient
    -0.64
     Maid
    -0.62
    ion
    -0.61
     reservoirs
    -0.60
     Parade
    -0.59
     Relief
    -0.59
    POSITIVE LOGITS
     narrowly
    1.27
     minded
    0.87
    velength
    0.87
     aver
    0.82
    iffe
    0.81
    Ń·
    0.80
     tailored
    0.77
     exting
    0.75
    eteen
    0.75
    byss
    0.75
    Activations Density 0.002%

    No Known Activations